Safety monitoring system of liquid cooling over-charging pile

Abstract

The invention discloses a safety monitoring system of a liquid cooling over-charging pile, and relates to the technical field of over-charging pile monitoring, the system comprises a data acquisition module, a data processing and analysis module, a dynamic model construction module, a temperature prediction module and a safety early warning module; according to the method, the dynamic model is constructed through the long short-term memory network LSTM, the complex nonlinear relation and the time sequence dependence of the multi-dimensional data are mined by using the gating mechanism of the dynamic model, the accurate characterization of the operation state of the liquid cooling over-charging pile is realized, the actual operation state of the equipment can be accurately described, the temperature data time sequence modeling is performed through the LSTM, and the accuracy of the temperature data time sequence modeling is improved. Parameters such as multi-source temperature and cooling liquid flow are fused, real-time prediction of the temperature change trend is achieved, the defect that a traditional algorithm is insufficient in temperature time sequence dependence capture is overcome, temperature abnormity can be recognized in advance, a safety threshold value is dynamically adjusted through a fuzzy logic algorithm, and self-adaptive threshold value adjustment is achieved in combination with parameters such as charging power. The problem that a traditional fixed threshold value is poor in adaptability is solved, and the early warning accuracy is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of supercharging pile monitoring, and in particular to a safety monitoring system for a liquid-cooled supercharging pile. Background Art

[0002] A liquid-cooled supercharging pile is a device that uses liquid cooling technology to achieve high-power charging. It can quickly recharge new energy vehicles and other terminals in a short period of time. Its core is to control the temperature during the charging process through coolant circulation, solving the heat dissipation bottleneck of traditional air cooling technology in supercharging scenarios. It is a key infrastructure in the current new energy charging field. Since liquid-cooled supercharging piles generate a lot of heat when running at high power, if the temperature control fails, it may cause cable overheating, battery thermal runaway and even fire and other safety risks. Therefore, a real-time monitoring system is required to dynamically monitor multi-dimensional parameters such as temperature, power, and current during the charging process to ensure that the equipment operates within the safety threshold.

[0003] According to the patent application number 202311291612.6, a safety monitoring system for liquid-cooled supercharging technology is disclosed, including: collecting usage data of several charging piles and charging piles to be tested; obtaining reference charging piles and non-reference charging piles, and obtaining the aging degree of each usage process for the reference charging pile; obtaining correction weight values ​​between the usage processes of different reference charging piles, matching the aging degrees of different reference charging piles according to the correction weight values, and obtaining a standard aging change curve; obtaining the aging degree of each usage process of each non-reference charging pile, clustering the usage processes according to the aging degree, and completing the construction of a dynamic model of the temperature of the charging pile to be tested; completing temperature monitoring and early warning of the charging pile to be tested by performing Kalman filtering on the dynamic model. The above solution solves the problem that in the process of monitoring the device temperature through Kalman filtering, too many factors affecting the device lead to inaccurate dynamic model and inaccurate monitoring results.

[0004] However, the above scheme still has certain defects when used. The dynamic model of the above scheme mainly relies on algorithms such as Kalman filtering, which makes it difficult to capture the complex nonlinear coupling relationship between multi-dimensional data such as charging current, voltage, and temperature, resulting in the model being unable to accurately describe the actual operating status of the equipment, and risk warning is delayed. Temperature prediction mostly relies on algorithms such as Kalman filtering, which have insufficient modeling capabilities for the time series dependence of temperature data and cannot effectively predict temperature change trends, resulting in warning lags and difficulty in identifying temperature anomalies in advance. Therefore, it is of great significance to develop a safety monitoring system for liquid-cooled supercharging piles. Summary of the Invention

[0005] The purpose of the present invention is to make up for the shortcomings of the existing technology and provide a safety monitoring system for liquid-cooled supercharging piles. It can build a dynamic model through the long short-term memory network LSTM, use its gating mechanism to mine the complex nonlinear relationships and time series dependencies of multi-dimensional data, and realize accurate characterization of the operating status of the liquid-cooled supercharging pile. It can accurately describe the actual operating status of the equipment, model the temperature data time series through LSTM, integrate multi-source temperature and coolant flow parameters, and realize real-time prediction of temperature change trends.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: a safety monitoring system for liquid-cooled supercharging piles, the system comprising: a data acquisition module, a data processing and analysis module, a dynamic model building module, a temperature prediction module and a safety warning module; The data acquisition module collects multi-dimensional data of the liquid-cooled supercharging pile during charging in real time through current sensors, voltage sensors, flow sensors, temperature sensors and battery management systems. The multi-dimensional data includes charging current, charging voltage, charging time, coolant flow, coolant temperature, pile surface temperature, ambient temperature and battery status parameters; The data processing and analysis module pre-processes the collected multi-dimensional data and extracts data features; The dynamic model building module uses a long short-term memory (LSTM) network to build a dynamic model of the liquid-cooled supercharging pile based on preprocessed multi-dimensional data and extracted features. The LSTM captures complex nonlinear relationships and long-term temporal dependency information in the data through a gating mechanism. The temperature prediction module uses the LSTM algorithm to perform time series modeling on the temperature data including the coolant temperature, pile surface temperature and ambient temperature to achieve real-time temperature prediction; The safety warning module compares the operating status data and temperature prediction data predicted by the dynamic model with the preset safety thresholds, and triggers sound and light alarms, SMS notifications, and warning signals pushed by the APP when the thresholds are exceeded.

[0007] Furthermore, in the data acquisition module, the current sensor adopts a Hall current sensor, the voltage sensor adopts a voltage divider voltage sensor, the flow sensor adopts an electromagnetic flowmeter, and the temperature sensor adopts a thermocouple sensor. The battery management system communicates with the data acquisition module through the CAN bus, and the communication protocol adopts a standardized communication protocol.

[0008] Furthermore, the data processing and analysis module preprocesses the collected multi-dimensional data, including data cleaning and normalization operations. When the data processing and analysis module performs feature extraction, it first extracts the basic features of the time domain and frequency domain of the preprocessed multi-dimensional data, and then generates a feature vector that integrates the multi-dimensional information through correlation analysis, dimensionality reduction and composite feature construction operations.

[0009] Furthermore, the LSTM network of the dynamic model building module adopts an improved gate control structure, including an input gate, a forget gate, an output gate and a newly added temperature perception gate. The temperature perception gate formula is: ,in, is the temperature sensing gate output, is the Sigmoid function, , is the weight matrix, , is the bias vector, which is determined by iterative optimization in historical multi-dimensional data training through the back propagation algorithm. is the temperature feature vector at the current moment. By integrating temperature data, the model’s sensitivity to thermal runaway risk is enhanced. Indicates that the LSTM network is The hidden layer output at time t.

[0010] Furthermore, the temperature prediction module adopts a multivariate time series prediction model, the input includes coolant temperature, pile surface temperature, ambient temperature and coolant flow, and the output is the predicted value of the pile core temperature at a future time. The prediction formula is: ,in, Output activation function for the LSTM hidden layer, , is the time series feature weight, which is automatically learned during model training through the attention mechanism and is used to characterize the contribution of input parameters at different time steps to temperature prediction. is the sampling interval, is the noise term, and the prediction accuracy is improved by fusing multi-physics field parameters.

[0011] Furthermore, the threshold dynamic adjustment algorithm of the safety warning module is based on fuzzy logic control. The input is the current charging power, battery SOC status and historical warning frequency, and the output is the temperature warning threshold. The calculation formula is: in, is the basic threshold, is the fuzzy rule weight coefficient, which is initialized by the expert experience parameters of the fuzzy rule base and iteratively updated by the Bayesian optimization algorithm based on historical warning data. It is the ambient temperature compensation item, which enables the threshold to be adaptively adjusted according to the operating conditions.

[0012] Furthermore, the system also includes a data storage module that uses a time-series database to store data and supports batch queries indexed by charging cycle. The storage structure includes device ID, timestamp, original data field and model prediction result field, and the data retention period meets industry regulatory requirements.

[0013] Furthermore, the system also includes a fault diagnosis module, which constructs a residual sequence based on the dynamic model prediction value and the actual monitoring value, detects system anomalies through the residual threshold, uses the fault tree analysis method FTA to establish a mapping relationship between the fault cause and the symptoms, realizes fault location, configures a fault severity assessment model, calculates the risk index according to the fault type, duration and impact range, and generates a fault handling plan, including an emergency shutdown strategy, a power derating curve and a maintenance guidance process.

[0014] Compared with existing technologies, the safety monitoring system of the liquid-cooled supercharging pile has the following beneficial effects: The present invention constructs a dynamic model through the long short-term memory network LSTM, and uses its gating mechanism to mine the complex nonlinear relationships and time series dependencies of multi-dimensional data to achieve accurate characterization of the operating status of the liquid-cooled supercharging pile. It can accurately describe the actual operating status of the equipment, model the temperature data time series through LSTM, integrate multi-source temperature and coolant flow parameters, and realize real-time prediction of temperature change trends, overcoming the defect of traditional algorithms that cannot capture the temperature time series dependency. It can identify temperature anomalies in advance, dynamically adjust the safety threshold through fuzzy logic algorithm, and realize adaptive adjustment of threshold in combination with parameters such as charging power, thus solving the problem of poor adaptability of traditional fixed threshold and improving the accuracy of early warning.

[0015] Other advantages, objects and features of the present invention will be described in part in the following description and, in part, will be apparent to those skilled in the art based on an examination of the following or may be learned from the practice of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] To more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. Those skilled in the art can also derive other drawings based on these drawings without inventive effort.

[0017] Figure 1 This is a schematic diagram of the structure of the safety monitoring system of the liquid-cooled supercharging pile; Figure 2 This is a schematic diagram of the working process of the safety monitoring system of the liquid-cooled supercharging pile. DETAILED DESCRIPTION

[0018] In order to further illustrate the technical means and effects adopted by the present invention to achieve the predetermined purpose of the invention, the specific implementation methods, structures, features and effects of the present invention are described in detail below in conjunction with the accompanying drawings and preferred embodiments. Example 1

[0019] At a busy new energy vehicle charging station, an electric car drives in and connects to a liquid-cooled supercharging pile for charging. This charging station serves many vehicles daily, with high charging power and a complex environment, placing extremely high demands on the safe and stable operation of the supercharging pile. Figure 1 and Figure 2 The liquid-cooled supercharging pile safety monitoring system plays a key role in this scenario. The specific implementation process is as follows: The data acquisition module starts working first. The Hall current sensor collects the charging current in real time, the voltage divider sensor collects the charging voltage, the electromagnetic flowmeter collects the coolant flow, and the thermocouple sensor collects the coolant temperature, pile surface temperature and ambient temperature. The battery management system transmits the battery status parameters to the data acquisition module through the CAN bus, thereby obtaining multi-dimensional data.

[0020] The collected data is transmitted to the data processing and analysis module, where data cleaning is first performed. Data mutation points are detected through a sliding window, and nonlinear compensation correction is performed on data points that exceed the historical statistical interval. The correction amount is exponentially related to the degree of deviation. Then, the scaling factor is dynamically adjusted based on the data distribution characteristics. The scaling factor is adaptively calculated according to the interquartile range of the data to complete the normalization operation. Finally, the time-frequency feature matrix of the data is extracted through a joint time-frequency domain transformation, and the multi-resolution analysis method is used to decompose signal features of different scales to realize data feature extraction.

[0021] The dynamic model building module receives the processed data and uses the long short-term memory network (LSTM) for deep mining. The LSTM network uses an improved gate control structure, including an input gate, a forget gate, an output gate, and a newly added temperature perception gate. The temperature perception gate formula is: ,in, is the temperature sensing gate output, is the Sigmoid function, , is the weight matrix, , is the bias vector, which is determined by iterative optimization in historical multi-dimensional data training through the back propagation algorithm. The temperature feature vector at the current moment is used to enhance the model's sensitivity to thermal runaway risks by integrating temperature data. The model is used to predict the future operating status of the supercharging pile, such as power changes during charging and changes in the battery's charging status.

[0022] The temperature prediction module uses a multivariate time series prediction model to process temperature data. The input includes coolant temperature, pile surface temperature, ambient temperature and coolant flow rate. The output is the predicted value of the pile core temperature at the future moment. The prediction formula is: ,in, Output activation function for the LSTM hidden layer, , is the time series feature weight, which is automatically learned during model training through the attention mechanism and is used to characterize the contribution of input parameters at different time steps to temperature prediction. is the sampling interval, is the noise term, and the prediction accuracy is improved by fusing multi-physics field parameters.

[0023] The safety warning module compares the operating status data and temperature prediction data predicted by the dynamic model with the preset safety threshold. The threshold dynamic adjustment algorithm is based on fuzzy logic control. The input is the current charging power, battery SOC status and historical warning frequency. The output is the temperature warning threshold. The calculation formula is: in, is the basic threshold, is the fuzzy rule weight coefficient, which is initialized by the expert experience parameters of the fuzzy rule base and iteratively updated by the Bayesian optimization algorithm based on historical warning data. It is an ambient temperature compensation item, which realizes adaptive adjustment of the threshold according to the operating conditions. When the predicted result exceeds the safety threshold, an early warning signal is immediately issued through sound and light alarms, SMS notifications, APP push, etc.

[0024] If the system detects an anomaly, the fault diagnosis module constructs a residual sequence based on the dynamic model prediction value and the actual monitoring value, and detects system anomalies through the residual threshold; uses the fault tree analysis method (FTA) to establish a mapping relationship between the fault cause and symptoms to achieve fault location; configures a fault severity assessment model to calculate the risk index based on the fault type, duration and impact range; and generates a fault handling plan, including an emergency shutdown strategy, power derating curve and maintenance guidance process.

[0025] To sum up, in this embodiment, through the full-process operation of the safety monitoring system, the shortcomings of traditional methods in nonlinear relationship capture, time-dependent modeling and threshold adaptive adjustment are effectively overcome, and accurate monitoring and prediction of the operating status and temperature of liquid-cooled supercharging piles are achieved. It can quickly locate faults and provide disposal solutions, significantly improving the safety performance and operational reliability of supercharging piles in complex application scenarios, and ensuring the stable and efficient operation of charging stations. Example 2

[0026] At a charging station in a highway service area, an electric truck that has been traveling for a long distance stopped and connected to a liquid-cooled supercharging pile for recharging. The service area is located in a high-temperature and high-humidity area, and the electric truck has a large charging power demand and a concentrated charging time, which places more stringent requirements on the heat dissipation performance of the supercharging pile and the real-time performance of the safety monitoring system. Figure 1 and Figure 2 The specific implementation process of the liquid-cooled supercharging pile safety monitoring system in this scenario is as follows: First, the Hall current sensor in the data acquisition module continuously monitors the pulse current fluctuations in the charging circuit, the voltage divider voltage sensor tracks the dynamic changes of the charging voltage in real time, the electromagnetic flowmeter synchronously collects real-time flow data of the coolant, and the thermocouple sensor performs high-frequency sampling of the coolant temperature, pile surface temperature, and ambient temperature and humidity. At the same time, the battery management system of the electric truck transmits parameters such as the battery SOC status and internal resistance changes to the data acquisition module in real time through the CAN bus, forming an original data set containing multi-dimensional physical quantities.

[0027] The collected raw data first enters the data cleaning stage. The sliding median filter algorithm is used to remove spike noise in the temperature data, and the linear interpolation method is used to fill the short-term missing values ​​of the flow data. During the normalization process, according to the parameter characteristics of the electric truck charging scenario, the standard deviation normalization method is used for strong fluctuating signals such as voltage and current, and the normalization method is used for slowly varying signals such as temperature and humidity. In the feature extraction stage, the high-frequency harmonic components are separated from the current signal, and the multi-scale energy characteristics of the temperature signal are extracted using wavelet packet decomposition. The mutual information method is used to screen out feature combinations with a correlation with thermal runaway risk higher than the threshold, and generate a feature vector that integrates the spatiotemporal characteristics.

[0028] The improved LSTM network receives the preprocessed feature vector and enhances its sensitivity to high temperature anomalies through the newly added temperature perception gate. The temperature perception gate formula is: ,in, is the temperature sensing gate output, is the Sigmoid function, , is the weight matrix, , is the bias vector, which is determined by iterative optimization in historical multi-dimensional data training through the back propagation algorithm. The temperature feature vector at the current moment is used to enhance the model's sensitivity to thermal runaway risks by integrating temperature data. The model is used to predict the future operating status of the supercharging pile, such as power changes during charging and changes in the battery's charging status.

[0029] The temperature perception gate dynamically adjusts the gate weight by integrating feature vectors such as the surface temperature gradient of the pile and the temperature difference between the coolant inlet and outlet. This enables the model to capture early signals of local overheating during high-power charging of electric trucks. For example, when the temperature rise rate in a certain area of ​​the pile exceeds the historical average, the temperature perception gate will increase the input weight of the data in that dimension, driving the model to prioritize learning abnormal features, thereby predicting deviations from the operating status trend in advance.

[0030] The multivariable time series prediction model takes coolant temperature, ambient humidity, coolant flow rate, and battery charge rate as input, and outputs the predicted value of the pile core temperature at the future moment. The prediction formula is: ,in, Output activation function for the LSTM hidden layer, , is the time series feature weight, which is automatically learned during model training through the attention mechanism and is used to characterize the contribution of input parameters at different time steps to temperature prediction. is the sampling interval, The model improves the prediction accuracy by integrating multiple physical field parameters and automatically identifies key influencing factors through the attention mechanism. For example, in a high temperature and high humidity environment, the model will increase the weight of the influence of ambient humidity on the heat dissipation efficiency of the pile body. Combined with the historical fluctuation pattern of the coolant flow rate, the evolution trend of the core temperature of the pile body in the future period is predicted. When it is predicted that the core temperature may approach the warning threshold within a certain charging cycle, the system triggers the coolant pump power adjustment logic in advance to prevent temperature exceeding the limit by dynamically increasing the flow rate.

[0031] The full warning module compares the operating status data and temperature prediction data predicted by the dynamic model with the preset safety threshold. The threshold dynamic adjustment algorithm is based on fuzzy logic control. The input is the current charging power, battery SOC status and historical warning frequency. The output is the temperature warning threshold. The calculation formula is: , the threshold is adaptively adjusted according to the operating conditions. When the predicted result exceeds the safety threshold, an early warning signal is immediately issued through sound and light alarms, SMS notifications, APP push, etc.

[0032] If the warning is continuously triggered, the fault diagnosis module initiates residual analysis: by comparing the coolant flow predicted by the LSTM model with the actual monitored value, a residual sequence is constructed and its root mean square error (RMSE) is calculated. When the RMSE exceeds the preset threshold, it is determined to be a "cooling system abnormality" and the fault tree analysis method is used to locate the specific cause of the fault, such as "coolant pump impeller wear". At the same time, a disposal plan is generated that includes a derating charging strategy (such as reducing the power to 70% of the rated value) and maintenance priority, ensuring that risk control is completed without interrupting the charging of critical vehicles.

[0033] All real-time data and model prediction results are stored by charging cycle in a time-series database. The storage structure includes device ID, truck VIN code, timestamp, and multi-dimensional parameter fields. Operations and maintenance personnel can retrieve the full charging cycle data of a specific vehicle through the query interface, analyze the correlation between the charging power curve and temperature fluctuations in high-temperature environments, and provide data support for subsequent equipment heat dissipation design optimization.

[0034] In summary, in this embodiment, the system effectively copes with the complex working conditions of high-power charging of electric trucks in high-temperature and high-humidity environments through multi-dimensional data fusion, improved LSTM modeling and fuzzy logic threshold adjustment. Compared with the traditional fixed threshold monitoring solution, dynamic threshold adjustment improves the warning accuracy and shortens the fault location time. At the same time, by adjusting the heat dissipation strategy in advance, charging interruptions caused by temperature out of control are avoided, significantly improving the reliability and emergency response capabilities of supercharging piles in highway service areas.

[0035] The above description is merely a preferred embodiment of the present invention and does not constitute any form of limitation to the present invention. Although the present invention has been disclosed as above in terms of a preferred embodiment, it is not intended to limit the present invention. Any person skilled in the art can, without departing from the scope of the technical solution of the present invention, make some changes or modifications to equivalent embodiments using the technical contents disclosed above. However, any brief modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention are still within the scope of the technical solution of the present invention.

Claims

1. The safety monitoring system of liquid-cooled supercharging pile is characterized by: The system includes: data acquisition module, data processing and analysis module, dynamic model building module, temperature prediction module and safety warning module; The data acquisition module collects multi-dimensional data of the liquid-cooled supercharging pile during charging in real time through current sensors, voltage sensors, flow sensors, temperature sensors and battery management systems. The multi-dimensional data includes charging current, charging voltage, charging time, coolant flow, coolant temperature, pile surface temperature, ambient temperature and battery status parameters; The data processing and analysis module pre-processes the collected multi-dimensional data and extracts data features; The dynamic model building module uses a long short-term memory (LSTM) network to build a dynamic model of the liquid-cooled supercharging pile based on preprocessed multi-dimensional data and extracted features. The LSTM captures complex nonlinear relationships and long-term temporal dependency information in the data through a gating mechanism. The temperature prediction module uses the LSTM algorithm to perform time series modeling on the temperature data including the coolant temperature, pile surface temperature and ambient temperature to achieve real-time temperature prediction; The safety warning module compares the operating status data and temperature prediction data predicted by the dynamic model with the preset safety thresholds, and triggers sound and light alarms, SMS notifications, and warning signals pushed by the APP when the thresholds are exceeded.

2. The safety monitoring system for liquid-cooled supercharging piles according to claim 1, characterized in that: In the data acquisition module, the current sensor adopts a Hall current sensor, the voltage sensor adopts a voltage divider voltage sensor, the flow sensor adopts an electromagnetic flowmeter, and the temperature sensor adopts a thermocouple sensor. The battery management system communicates with the data acquisition module through the CAN bus, and the communication protocol adopts a standardized communication protocol.

3. The safety monitoring system for liquid-cooled supercharging piles according to claim 1, characterized in that: The data processing and analysis module preprocesses the collected multi-dimensional data, including data cleaning and normalization operations. When the data processing and analysis module performs feature extraction, it first extracts basic features in the time domain and frequency domain of the preprocessed multi-dimensional data, and then generates a feature vector that integrates multi-dimensional information through correlation analysis, dimensionality reduction and composite feature construction operations.

4. The safety monitoring system for liquid-cooled supercharging piles according to claim 1, characterized in that: The LSTM network of the dynamic model building module adopts an improved gate control structure, including an input gate, a forget gate, an output gate, and a newly added temperature perception gate. The temperature perception gate formula is: ,in, is the temperature sensing gate output, is the Sigmoid function, , is the weight matrix, , is the bias vector, is the temperature feature vector at the current moment, Indicates that the LSTM network is The hidden layer output at time t.

5. The safety monitoring system for liquid-cooled supercharging piles according to claim 1, characterized in that: The temperature prediction module adopts a multivariate time series prediction model. The input includes coolant temperature, pile surface temperature, ambient temperature and coolant flow rate. The output is the predicted value of the pile core temperature at the future time. The prediction formula is: ,in, Output activation function for the LSTM hidden layer, , is the time series feature weight, is the sampling interval, is the noise term.

6. The safety monitoring system for liquid-cooled supercharging piles according to claim 1, characterized in that: The threshold dynamic adjustment algorithm of the safety warning module is based on fuzzy logic control. The input is the current charging power, battery SOC status and historical warning frequency. The output is the temperature warning threshold. The calculation formula is: in, is the basic threshold, is the fuzzy rule weight coefficient, It is the ambient temperature compensation item, which enables the threshold to be adaptively adjusted according to the operating conditions.

7. The safety monitoring system for liquid-cooled supercharging piles according to claim 1, characterized in that: The system also includes a data storage module that uses a time series database to store data and supports batch queries indexed by charging cycle. The storage structure includes device ID, timestamp, original data field and model prediction result field.

8. The safety monitoring system for liquid-cooled supercharging piles according to claim 1, characterized in that: The system also includes a fault diagnosis module, which constructs a residual sequence based on the dynamic model prediction value and the actual monitoring value, detects system anomalies through the residual threshold, uses the fault tree analysis method FTA to establish a mapping relationship between the fault cause and the symptom to achieve fault location, configures a fault severity assessment model, calculates the risk index according to the fault type, duration and impact range, and generates a fault handling plan, including an emergency shutdown strategy, a power derating curve and a maintenance guidance process.

Images