Inverse modeling methods for indoor airborne pollutant tracking: literature review and fundamentals

Reduction in indoor environment quality calls for effective control and improvement measures. Accurate and prompt identification of contaminant sources ensures that they can be quickly removed and contaminated spaces isolated and cleaned. This paper discusses the use of inverse modeling to identify potential indoor pollutant sources with limited pollutant sensor data. The study reviews various inverse modeling methods for advection-dispersion problems and summarizes the methods into three major categories: forward, backward, and probability inverse modeling methods. The adjoint probability inverse modeling method is indicated as an appropriate model for indoor air pollutant tracking because it can quickly find source location, strength and release time without prior information. The paper introduces the principles of the adjoint probability method and establishes the corresponding adjoint equations for both multi-zone airflow models and computational fluid dynamics (CFD) models. The study proposes a two-stage inverse modeling approach integrating both multi-zone and CFD models, which can provide a rapid estimate of indoor pollution status and history for a whole building. Preliminary case study results indicate that the adjoint probability method is feasible for indoor pollutant inverse modeling. PRACTICAL IMPLICATIONS: The proposed method can help identify contaminant source characteristics (location and release time) with limited sensor outputs. This will ensure an effective and prompt execution of building management strategies and thus achieve a healthy and safe indoor environment. The method can also help design optimal sensor networks.     

Protecting a whole building from critical indoor contamination with optimal sensor network design and source identification methods

To maintain a healthful and secure indoor environment, it is crucial to design an effective indoor air quality (IAQ) sensor network and interpret sensor outputs for prompt IAQ controls. This paper introduces how a probability concept based inverse modeling method – the adjoint probability method – can be used to assist in designing a high-performance IAQ sensor network and identifying potential contaminant source locations for an entire building based on limited sensor outputs. The study proposes a new IAQ sensor network design and optimization method for buildings with one or more compartments on the basis of the probability calculation. With responses from optimized sensors, a two-stage integrated inverse prediction algorithm is developed that is able to identify a potential IAQ source zone (or room) in a building as well as an exact location within the room. The paper demonstrates the design of sensor networks and the application of the source identification algorithms for a residential dwelling. The case study verifies the feasibility, effectiveness and accuracy of the proposed sensor design method and the two-stage algorithm for indoor contaminant control.     

Principles and applications of probability-based inverse modeling method for finding indoor airborne contaminant sources

Building indoor air quality (IAQ) has received growing attentions lately because of the extended time people spend indoors and the increasing reports of health problems related to poor indoor environments. Recent alarms to potential terrorist attacks with airborne chemical and biological agents (CBA) have further highlighted the research needs on building vulnerability and protection. To maintain a healthful and safe indoor environment, it is crucial to identify contaminant source locations, strengths, and release histories. Accurate and prompt identification of contaminant sources can ensure that the contaminant sources can be quickly removed and contaminated spaces can be effectively isolated and cleaned. This paper introduces a probability concept based prediction method—the adjoint probability method-that can track potential indoor airborne contaminant sources with limited sensor outputs. The paper describes the principles of the method and presents the general modeling algorithm and procedure that can be implemented with current computational fluid dynamics (CFD) or multi-zone airflow models. The study demonstrates the application of the method for identifying airborne pollutant source locations in two realistic indoor environments with few sensor measurement outputs. The numerical simulations verify the feasibility and accuracy of the method for indoor pollutant tracking applications, which forms a good foundation for developing an intelligent and integrated indoor environment management system that can promptly respond to indoor pollution episodes with effective detection, analysis, and control.     

Prompt tracking of indoor airborne contaminant source location with probability-based inverse multi-zone modeling

Indoor air quality (IAQ) has a significant influence on occupants' comfort, health, productivity, and safety. Existing studies show that the primary causes of many IAQ problems are various airborne contaminants that either are generated indoors or penetrate into indoor environments with passive or active airflows. Accurate and prompt identification of contaminant sources can help determinate appropriate IAQ control solutions, such as, eliminating contaminant sources, isolating and cleaning contaminated spaces. This study develops a fast and effective inverse modeling method for identifying indoor contaminant source characteristics. The paper describes the principles of the probability-based adjoint inverse modeling method and formulates a multi-zone model based inverse prediction algorithm that can rapidly track contaminant source location with known source release time in a building with many compartments. The paper details the inverse modeling procedure with modification of an existing multi-zone airflow and contaminant transport simulation program. The application of the method has been demonstrated with two case studies: contaminant releases in a multi-compartment residential house and in a complex institutional building. The numerical experiments tested the source identification capability of the program for various contaminant sensing scenarios. The investigation verifies the effectiveness and accuracy of the developed method for indoor contaminant source tracking, which will be further explored to identify more complicated indoor contamination episodes.     

Inverse identification of multiple outdoor pollutant sources with a mobile sensor

Air pollution is becoming more and more severe in large cities. Accurate and rapid identification of outdoor pollutant sources can facilitate proper and effective air quality management in urban environments. Traditional “trial–error” process is time consuming and is incapacity in distinguishing multiple potential sources, which is common in urban pollution. Inverse prediction methods such as probability based adjoint modelling method have shown viability for locating indoor contaminant sources. This paper advances the adjoint probability method to track outdoor pollutant sources of constant release. The study develops an inverse modelling algorithm that can promptly locate multiple outdoor pollutant sources with limited pollution information detected by a movable sensor. Two numerical field experiments are conducted to illustrate and verify the predictions: one in an open space and the other in an urban environment. The developed algorithm promptly and accurately identifies the source locations in both cases. The requirement of an accurate urban building model is the primary prerequisite of the developed algorithm for urban application.     

辨识室内空气污染源的反问题建模

当发生室内空气污染事故时,获知污染源释放的位置与强度等信息至关重要.利用污染物传感器提供的信息来推断室内空气污染源的研究属于反问题建模.反问题属于病态问题,因而必须采取一些特定的策略才能让反问题获得求解.本文总结了国内外有关应用反问题建模来辨识室内空气污染源的研究进展,以及反问题建模在传热、水污染以及大气污染等领域内的研究概况.辨识室内空气污染源的研究方法可归纳为四类,即分析法、优化法、概率法以及直接求解法.直接求解法不需要使用过多的假设,而且能够较好平衡计算效率以及计算精度,因而比较适于室内环境中污染源的辨识.     

Identifying decaying contaminant source location in building HVAC system using the adjoint probability method

Building heating, ventilation and air-conditioning (HVAC) system can be potential contaminant emission source. Released contaminants from the mechanical system are transported through the HVAC system and thus impact indoor air quality (IAQ). Effective control and improvement measures require accurate identification and prompt removal of contaminant sources from the HVAC system so as to eliminate the unfavourable influence on the IAQ. This paper studies the application of the adjoint probability method for identifying a dynamic (decaying) contaminant source in building HVAC system. A limited number of contaminant sensors are used to detect contaminant concentration variations at certain locations of the HVAC ductwork. Using the sensor inputs, the research is able to trace back and find the source location. A multi-zone airflow model, CONTAM, is employed to obtain a steady state airflow field for the studied building with detailed duct network, upon which the adjoint probability based inverse tracking method is applied. The study reveals that the adjoint probability method can effectively identify the decaying contaminant source location in building HVAC system with few properly located contaminant concentration sensors.     

Experimental verification of tracking algorithm for dynamically-releasing single indoor contaminant

Identifying contaminant sources in a precise and rapid manner is critical to indoor air quality (IAQ) management as disclosed source information can facilitate proper and effective IAQ controls in environments with airborne infection, fire smoke and chemical pollutant release etc. Probability-based inverse modeling method was shown feasible for locating single instantaneous source in IAQ events. To tackle more realistic sources of continuous release, this paper advances the method to identify continuously releasing single contaminant source. The study formulates a suite of inverse modeling algorithms that can promptly locate dynamic source with known release time for IAQ events. Two field experiments are employed to verify the prediction: one in a multi-room apartment and the other in a hospital ward which was involved in a SARS outbreak in Hong Kong in 2003. The developed algorithms promptly and accurately identify the source locations in both cases.     

Identification of contaminant sources in enclosed spaces by a single sensor

To protect occupants from infectious diseases or possible chemical/biological agents released by a terrorist in an enclosed space, such as an airliner cabin, it is critical to identify gaseous contaminant source locations and strengths. This paper identified the source locations and strengths by solving inverse contaminant transport with the quasi-reversibility (QR) and pseudo-reversibility (PR) methods. The QR method replaces the second-order diffusion term in the contaminant transport equation with a fourth-order stabilization term. By using the airflow pattern calculated by computational fluid dynamics (CFD) and the time when the peak contaminant concentration was measured by a sensor in downstream, the QR method solves the backward probability density function (PDF) of contaminant source location. The PR method reverses the airflow calculated by CFD and solves the PDF in the same manner as the QR method. The position with the highest PDF is the location of the contaminant source. The source strength can be further determined by scaling the nominal contaminant concentration computed by CFD with the concentration measured by the sensor. By using a two-dimensional and a three-dimensional aircraft cabin as examples of enclosed spaces, the two methods can identify contaminant source locations and strengths in the cabins if the sensors are placed in the downstream location of the sources. The QR method performed slightly better than the PR method but with a longer computing time. PRACTICAL IMPLICATIONS: The paper presents a method that can be used to find a gaseous contaminant source location and determine its strength in enclosed spaces with the data of contaminant concentration measured by one sensor. The method can be a very useful tool to find where, what, and how the contamination has happened. The method is also useful for optimally placing sensors in enclosed spaces. The results can be applied to develop appropriate measures to protect occupants in enclosed environments from infectious diseases or chemical/biological warfare agents released by a terrorist.     

Locating time-varying contaminant sources in 3D indoor environments with three typical ventilation systems using a multi-robot active olfaction method

For a sudden contaminant release in an indoor environment, source localization can provide critical information for preventing and mitigating indoor air pollution and its related health and security problems. Considerable research has focused on locating indoor contaminant sources with instantaneous or constant release rates; however, few studies on locating indoor sources with time-varying release rates have been reported. This study proposed a multi-robot active olfactory method for promptly locating time-varying sources in 3D indoor environments. The method extends our previously proposed method for 2D indoor environments by redefining and reprogramming it in a 3D coordinate system and proposing a 3D source declaration algorithm. Via more than 200 numerical experiments in 3D indoor environments with mixing, displacement, and piston ventilation systems, the method was fully demonstrated and validated. The results show the applicability and reliability of the method and reveal the effects of space style, ventilation mode, source release rate, source location, and obstacle layout on source localization.     

Inverse identification of the release location,temporal rates,and sensor alarming time of an airborne pollutant source

With an accidental release of an airborne pollutant, it is always critical to know where, when, and how the pollutant has been released. Then, emergency measures can be scientifically advised to prevent any possible harm. This investigation proposes an inverse model to identify the release location, the temporal rate profile, and the sensor alarming time from the start of a pollutant release. The first step is to implement the inverse operation to the cause–effect matrix to obtain the release rate profiles for discrete candidate scenarios with concentration information provided by one sensor. The second step is to interpret the occurrence probability of each solution in the first step with the Bayesian model by matching the concentration at the other sensor. The proposed model was applied to identify a single pollutant source in a two‐dimensional enclosure using measurement data and in a three‐dimensional aircraft cabin with simulated data. The results show that the model is able to correctly determine the pollutant source location, the temporal rate profile, and the sensor alarming time. The known conditions for input into the inverse model include a steady flow field and the valid temporal concentrations at two different locations.     

双区域内离散污染源的模拟及分析

为研究污染源对非污染源区域的影响,具体考虑室内污染源、室外污染物浓度以及室内与室外通风方式的作用,对双区域内一般离散气态污染物在室内的浓度水平进行模拟和分析。通过对污染源的研究表明:室内污染源浓度的成倍增加会导致两个区域平均浓度均成倍增加,并且区域平均浓度是区域本身浓度与室外浓度的叠加值。通过对通风方式的研究表明:在污染源区域设置排气扇是有效的排污方法,有利于改善室内人居环境。     

室内可吸入空气品质的指标分析

来自建材中的挥发性有机物VOCs的释放和扩散对人体可吸入空气品质产生很大的影响。以地板上SBR板为研究对象,一维扩散方程为模型,计算了板内和板上表面的浓度值。结果表明:适度提高板的焙烤温度,可加快板内VOCs的释放。同时采用CFD方法对污染源处于不同位置下的室内浓度场进行分析,提出了衡量室内污染源位置对人体可吸入空气品质影响的评价指标:污染源分布率(CRP)。从人体可吸入化学污染物的角度分析,当地板为室内主要化学污染物时,房间内采用下进上出的置换通风方式,无助于室内空气品质的改善。     

Towards improved characterization of high-risk releases using heterogeneous indoor sensor systems

The sudden release of toxic contaminants that reach indoor spaces can be hazardous to building occupants. For an acutely toxic contaminant, the speed of the emergency response strongly influences the consequences to occupants. The design of a real-time sensor system is made challenging both by the urgency and complex nature of the event, and by the imperfect sensors and models available to describe it. In this research, we use Bayesian modeling to combine information from multiple types of sensors to improve the characterization of a release. We discuss conceptual and algorithmic considerations for selecting and fusing information from disparate sensors. To explore system performance, we use both real tracer gas data from experiments in a three-story building, along with synthetic data, including information from door-position sensors. The added information from door-position sensors is found to be useful for many scenarios, but not always. We discuss the physical conditions and design factors that affect these results, such as the influence of the door positions on contaminant transport. We highlight potential benefits of multisensor data fusion, challenges in realizing those benefits, and opportunities for further improvement.     

Identification of contaminant sources in enclosed environments by inverse CFD modeling

In case contaminants are found in enclosed environments such as aircraft cabins or buildings, it is useful to find the contaminant sources. One method to locate contaminant sources is by inverse computational fluid dynamics (CFD) modeling. As the inverse CFD modeling is ill posed, this paper has proposed to solve a quasi-reversibility (QR) equation for contaminant transport. The equation improves the numerical stability by replacing the second-order diffusion term with a fourth-order stabilization term in the governing equation of contaminant transport. In addition, a numerical scheme for solving the QR equation in unstructured meshes has been developed. This paper demonstrates how to use the inverse CFD model with the QR equation and numerical scheme to identify gaseous contaminant sources in a two-dimensional aircraft cabin and in a three-dimensional office. The inverse CFD model could identify the contaminant source locations but not very accurate contaminant source strength because of the dispersive property of the QR equation. The results also show that this method works better for convection dominant flows than the flows that convection is not so important. PRACTICAL IMPLICATIONS: This paper presents a methodology that can be used to find contaminant source locations and strengths in enclosed environments with the data of airflow and contaminants measured by sensors. The method can be a very useful tool to find where, what, and how contamination has happened. The results can be used to develop appropriate measures to protect occupants in the enclosed environments from infectious diseases or terrorist releases of chemical/biological warfare agents as well as to decontaminate the environments.     

Removal of contaminants released from room surfaces by displacement and mixing ventilation: modeling and validation

This paper presents the experimental and numerical modeling of contaminant dispersion in a full-scale environmental chamber with different room air distribution systems. For the experimental modeling, an area source with uniform emissions of a hypothetical contaminant (SF6) from the entire floor surface is designed and constructed. Two different types of ventilation are studied: displacement and mixing ventilation. A computer model for predicting the contaminant dispersion in indoor spaces was validated with experimental data. The validated model is used to study the effects of airflow and the area-source location on contaminant dispersion. Results show that the global airflow pattern has a strong impact on the distribution of the contaminants. In general, the personal exposure could be estimated by analyzing the relative source positions in the airflow pattern. Accordingly, the location of an exhaust diffuser may not greatly affect the airflow pattern, but can significantly affect the exposure level in the room. PRACTICAL IMPLICATIONS: When designing ventilation in addition to bringing fresh air to occupants, it is important to consider the removal of contaminants released in the off-gassing of building materials. Typical indoor off-gassing examples are emissions of volatile organic compounds from building enclosure surfaces such as flooring and painted walls. In this study, we conducted experimental and numerical modeling of different area sources in a mock-up office setup, with displacement or mixing ventilation. Displacement ventilation was as successful as mixing ventilation in removing the contaminant source from the floor area. Actually, the most important consideration in the removal of these contaminants is the relative position of the area source to the main airflow pattern and the occupied zone.     

Characterizing transportation of indoor gaseous contaminant using the state space method

This paper employs the state space method to characterize transportation of indoor gaseous pollutant in steady airflow field. From the differential equations governing contaminant transportation in space, the state space equation for transportation is proposed and the analytical solution is obtained. In the method, the matrix covering hologram of the transportation is derived. The state space equation is validated with the analytic solution for the case of the simultaneous transportation of the pollution for piston flow. Similarly, the concentration from the proposed method for a 2-D case also agrees well with the result from CFD method based on the experimentally validated flow field. Based upon the analytic solution of the state equation, it is easily known that the influence of the initial concentration distribution and the pollution source on the concentration at the specific point. In addition, assisted by Chen’s zero equation turbulence model [1], the concentration field for a 3-D case is simulated by the presented method. It is found that there exists a regular stage at which the relative effect of the initial concentration distribution and the source on the concentration field will not change with time.     

Contaminant dispersion with personal displacement ventilation,Part I: Base case study

Personal displacement ventilation (PDV) is a new ventilation concept that combines the positive features of displacement ventilation with those of task conditioning or personalized ventilation. PDV is expected to create a micro-environment around an occupant to control the environment individually. In this study, a base PDV case with a contaminant source at different locations was modeled for contaminant dispersion in a full-scale chamber. Computational fluid dynamics (CFD) was used to simulate the indoor airflow and pollutant transport, and the simulation results were validated against the experimental data. The contaminant concentration field for three different contaminant source locations was analyzed. Based on our results, it seems that this kind of PDV system cannot create the expected “micro-environment” to avoid the disturbance of the outside airflow. Further studies on how to improve the PDV performance are given in the companion paper.     

Multi-objective optimization for sensor placement against suddenly released contaminant in air duct system

When a chemical or biological agent is suddenly released into a ventilation system, its dispersion needs to be promptly and accurately detected. In this work, an optimization method for sensors layout in air ductwork was presented. Three optimal objectives were defined, i.e. the minimum detection time, minimum contaminant exposure, and minimum probability of undetected pollution events. Genetic algorithm (GA) method was used to obtain the non-dominated solutions of multiobjectives optimization problem and the global optimal solution was selected among all of the non-dominated solutions by ordering solutions method. Since the biochemical attack occurred in a ventilation system was a random process, two releasing scenarios were proposed, i.e. the uniform and the air volume-based probability distribution. It was found that such a probability distribution affected the results of optimal sensors layout and also resulted in different detect time and different probability of undetected events. It was discussed how the objective functions are being compatible and competitive with each other, and how sensor quantity affect the optimal results and computational load. The impact of changes on other parameters was given, i.e. the deposition coefficient, the air volume distribution and the manual releasing. This work presents an angle of air ductwork design for indoor environment protection and expects to help in realizing the optimized sensor system design for sudden contaminant releasing within ventilation systems.     

室内空气环境的数值研究方法

建筑室内空气环境(IAE)与人们的舒适、健康及工作效率密切相关，对其数值模拟与评价具有重要的实际意义。由于室内存在大量离散分布的热源与污染源，IAE表现为复杂的对流传热传质过程，如何有效地利用CFD模拟技术分析与评价IAE即是本文的研究内容。作者根据室内空气环境的特殊性提出了对流传输与自然模拟方法，简洁有效地分析了二维层流置换通风房间中离散热与污染源之间的相互作用及其对IAE的影响。