A monitoring method and device for geotechnical energy storage systems
By acquiring multiple types of monitoring signals and performing targeted compensation, combined with state prediction models and databases, the signal deviation problem caused by the complexity of factors in the monitoring of geotechnical energy storage systems was solved, enabling accurate judgment of system state and resource optimization.
Patent Information
- Application Number
- CN202611515003.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-09-28
- Publication Date
- 2026-11-17
AI Technical Summary
Existing monitoring methods for geotechnical energy storage systems are unable to fully consider complex environmental factors, leading to deviations in monitoring signals, failure to accurately reflect the true state of the system, and lack of dynamic adjustment capabilities, which affects the accuracy and safety of system state prediction.
By acquiring multiple types of monitoring signals, compensating based on different target influencing factors, and utilizing state prediction models and preset feature databases, the monitoring frequency and area are dynamically adjusted to achieve standardized judgment of system state.
It improves the accuracy and reliability of monitoring data, enables early prediction of system operation trends, reduces resource consumption, and ensures the authenticity and precision of monitoring results.
Smart Images

Figure CN123111376A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of geotechnical energy storage technology, and in particular to a monitoring method and device for geotechnical energy storage systems. Background Technology
[0002] With the increasing global demand for clean energy, geotechnical energy storage systems, as a highly efficient and environmentally friendly energy storage method, are being used more and more widely in the energy sector. Geotechnical energy storage systems utilize the thermal and gas storage properties of soil and rock media to achieve efficient energy storage and release, which is of great significance for alleviating the energy supply and demand imbalance and promoting sustainable energy development.
[0003] However, the operating environment of geotechnical energy storage systems is complex and variable. Numerous factors, including soil and rock properties, ambient temperature, humidity, and pressure, significantly impact their performance and stability. Specifically, different monitoring signals, such as temperature, pressure, displacement, and humidity, are influenced by different dominant factors, and using a uniform compensation method can easily lead to signal deviations. Furthermore, existing state assessments often rely on single thresholds, lacking standardized matching based on state feature databases, and monitoring frequencies are typically fixed, making dynamic adjustments based on health status and environmental changes difficult. Existing monitoring methods often fail to comprehensively consider these complex factors, resulting in biased monitoring signals that cannot accurately reflect the system's true operating state. This leads to insufficient accuracy in system state prediction, hindering the timely detection of potential faults and safety hazards.
[0004] Therefore, there is an urgent need for a monitoring method and device for geotechnical energy storage systems to improve the accuracy and reliability of monitoring. Summary of the Invention
[0005] To address the aforementioned technical problems, this application provides a monitoring method and apparatus for geotechnical energy storage systems.
[0006] A first aspect of this application provides a monitoring method for a geotechnical energy storage system, comprising: Acquire various monitoring signals from the target geotechnical energy storage system; Each type of monitoring signal is compensated based on the target influencing factors to obtain multiple types of target monitoring signals; Among them, the target influencing factors corresponding to each type of monitoring signal are different. The target influencing factors are selected from the set of preset influencing factors that affect the performance of the target geotechnical energy storage system. Multiple target monitoring signals are input into the state prediction model for processing to obtain the predicted state data of the target geotechnical energy storage system. The target state of the target geotechnical energy storage system is determined based on the predicted state data and the pre-set state feature database.
[0007] A second aspect of this application provides a monitoring device for a geotechnical energy storage system, comprising: The signal acquisition module is used to acquire various monitoring signals of the target geotechnical energy storage system; The signal compensation module is used to compensate for each type of monitoring signal based on the target influencing factors, so as to obtain multiple types of target monitoring signals; Among them, the target influencing factors corresponding to each type of monitoring signal are different. The target influencing factors are selected from the set of preset influencing factors that affect the performance of the target geotechnical energy storage system. The data processing module is used to input multiple types of target monitoring signals into the state prediction model for processing, and obtain the predicted state data of the target geotechnical energy storage system. The state determination module is used to determine the target state of the target geotechnical energy storage system based on the predicted state data and the preset state feature database.
[0008] A third aspect of this application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor executes the computer program to implement the steps of the above-described monitoring method for a geotechnical energy storage system.
[0009] In a fourth aspect of this application, a computer-readable storage medium is provided, which stores a computer program that, when executed by a processor, implements the steps of the monitoring method for a geotechnical energy storage system described above.
[0010] The beneficial effects of the monitoring method and device for geotechnical energy storage systems provided in this application are as follows: Firstly, by acquiring multiple types of monitoring signals, this application can comprehensively capture various information during the operation of the geotechnical energy storage system, avoiding the bias and limitations of a single signal and providing a rich data foundation for system status assessment. Secondly, targeted compensation of monitoring signals based on different target influencing factors effectively eliminates interference from environmental changes, equipment aging, and other factors, improving data accuracy and reliability, and making the monitoring results more realistically reflect the actual system status. Then, by processing the compensated target monitoring signals using a state prediction model, the future operating trend of the system can be predicted in advance. Combined with a preset state feature database, the target state of the system can be determined, achieving standardization and accurate judgment of monitoring results. Furthermore, by calculating a health index based on the target state and dynamically adjusting the monitoring frequency and key monitoring areas accordingly, monitoring resource consumption can be reduced while ensuring monitoring accuracy. Attached Figure Description
[0011] Figure 1 A flowchart illustrating a monitoring method for a geotechnical energy storage system provided in an embodiment of this application; Figure 2 A structural block diagram of a monitoring device for a geotechnical energy storage system provided in an embodiment of this application; Figure 3 This is a schematic block diagram of an electronic device provided in an embodiment of this application. Detailed Implementation
[0012] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.
[0013] To make the purpose, technical solution, and advantages of this application clearer, the following will be described in conjunction with the appendix. Figure 1-3 The following is an explanation using specific examples.
[0014] Please refer to Figure 1 , Figure 1 A flowchart illustrating a monitoring method for a geotechnical energy storage system provided in an embodiment of this application is shown. The method includes: S101: Acquire various monitoring signals of the target geotechnical energy storage system.
[0015] In this embodiment, multiple monitoring signals of the target geotechnical energy storage system are acquired through at least two of the following sensing devices: temperature sensor, pressure sensor, displacement sensor, and humidity sensor. These sensors comprehensively collect temperature field signals, pressure distribution signals, geotechnical displacement signals, and humidity signals. Specifically, a distributed fiber optic temperature sensor is used, a vibrating wire pressure sensor is used, a high-precision laser displacement sensor is used, and a capacitive humidity sensor is used. The sampling frequency for each sensor is set to 1 time per minute. After acquiring the multiple monitoring signals of the target geotechnical energy storage system, this embodiment removes noise and interference from the monitoring signals to improve signal quality. Furthermore, monitoring signals of different types and magnitudes are normalized to facilitate subsequent analysis and comparison.
[0016] S102: Compensate each type of monitoring signal based on the target influencing factors to obtain multiple types of target monitoring signals; among them, the target influencing factors corresponding to each type of monitoring signal are different, and the target influencing factors are selected from the preset set of influencing factors that affect the performance of the target geotechnical energy storage system.
[0017] In this embodiment, an influencing factor analysis mechanism is established. Combining historical operational data, geological survey reports, and meteorological data, target influencing factors are selected for each type of monitoring signal from a pre-set set of influencing factors; and at least two types of monitoring signals correspond to different target influencing factors. For example, temperature signals are affected by ambient air temperature and geothermal gradients; pressure signals are affected by soil and rock stress and fluid pressure, etc. For existing typical influencing factor-signal relationships, signal compensation models are directly matched from a pre-set compensation model library; for complex or newly emerging influencing factors, by analyzing the first monitoring signal set (original monitoring signal data) and the second monitoring signal set (monitoring signal data obtained under ideal conditions), machine learning algorithms are used to determine compensation factors, thereby training an appropriate signal compensation model to complete the compensation for each type of monitoring signal and obtain multiple types of target monitoring signals.
[0018] Specifically, for the temperature signal, the target influencing factor is determined to be the fluctuation of ambient temperature. Historical ambient temperature data is selected from the preset set of influencing factors, and a compensation model between ambient temperature and soil temperature is established using a linear regression algorithm. By substituting the current ambient temperature data into the compensation model, the original temperature signal is corrected to obtain the target temperature signal. For the pressure signal, the target influencing factor is the change in the self-weight stress of the soil and rock mass. The density distribution data of the soil and rock mass and the stratum depth data are obtained from the preset influencing factor set. The pressure compensation model is constructed using the finite element analysis method. The original pressure signal is compensated according to the calculation results of the model to obtain the target pressure signal. For the displacement signal, the target influencing factor is seasonal freeze-thaw action. Historical freeze-thaw cycle data and frozen soil depth data are extracted from the preset influencing factor set. A displacement compensation model is established using a time series analysis algorithm. The original displacement signal is adjusted based on the model output to obtain the target displacement signal. For the humidity signal, the target influencing factor is groundwater seepage. Groundwater level change data and permeability coefficient data are retrieved from the preset influencing factor set. A humidity compensation model is constructed using Darcy's law. The original humidity signal is corrected based on the model calculation value to obtain the target humidity signal.
[0019] S103: Input the monitoring signals of multiple targets into the state prediction model for processing to obtain the predicted state data of the target geotechnical energy storage system.
[0020] In this embodiment, the predicted state data is used to characterize the current state and / or predict future states. The state prediction model adopts a fusion structure of a one-dimensional convolutional neural network and a long short-term memory network, including an input layer, a convolutional feature extraction layer, an LSTM temporal modeling layer, and a fully connected output layer. The input consists of multi-class target monitoring signals spliced together in time series, with an input dimension of N×T×C, where N is the number of samples, T is the time window length, and C is the number of monitoring signal categories; the output is the predicted state data, including at least one of temperature change rate, pressure fluctuation amplitude, displacement rate, humidity change trend, and fault risk probability. The training sample size is no less than 1000 groups, with each group having a time window of 24 hours and a sampling interval of 1 minute; of which at least 800 groups are normal operation samples, and warning and fault samples are supplemented through historical records, maintenance records, and simulation data. The loss function is mean squared error, the optimizer is Adam, the initial learning rate is set to 0.001, the number of training epochs is set to 200, and the early stop patience value is set to 20. Evaluation metrics include root mean square error, mean absolute error, and state recognition accuracy. Training stops when the root mean square error of the validation set no longer decreases for 20 consecutive rounds, and the trained state prediction model is obtained.
[0021] S104: Determine the target state of the target geotechnical energy storage system based on the predicted state data and the preset state feature database.
[0022] In this embodiment, cosine similarity is calculated between the predicted state data and each state feature vector in the state feature database. The state category corresponding to the state feature vector with the highest similarity is selected as the target state of the target geotechnical energy storage system. The state feature database includes four state categories: normal operation, early warning, fault, and unknown. Sample sources include historical monitoring data, maintenance records, fault waveform data, simulation data, and expert-annotated data. The construction method of the state feature database includes: extracting temperature, pressure, displacement, and humidity features from historical samples, normalizing them, and then clustering them using the K-means clustering algorithm. The number of clusters K is set to 4, and the optimal K value is determined by the silhouette coefficient. Each cluster center is used as the state feature vector of the corresponding state category. Cosine similarity is calculated between the predicted state data and each state feature vector in the state feature database to obtain multiple similarity values. The preset similarity threshold is 0.85. When the maximum similarity value is greater than or equal to 0.85, the corresponding state category is determined as the target state; when the maximum similarity value is less than 0.85, the target geotechnical energy storage system is determined to be in an unknown state, and an anomaly early warning mechanism is triggered.
[0023] As can be seen from the above, this application, by acquiring multiple types of monitoring signals, can comprehensively capture various information during the operation of the geotechnical energy storage system, avoiding the one-sidedness and limitations of a single signal, and providing a rich data foundation for system status assessment. Secondly, by specifically compensating the monitoring signals based on different target influencing factors, interference from factors such as environmental changes and equipment aging is effectively eliminated, improving data accuracy and reliability, and making the monitoring results more realistically reflect the actual state of the system. Subsequently, the compensated target monitoring signals are processed using a state prediction model, which can predict the future operating trend of the system in advance. Combined with a pre-set state characteristic database, the target state of the system is determined, achieving standardization and accurate judgment of the monitoring results.
[0024] In one embodiment of this application, compensation is performed on each type of monitoring signal based on target influencing factors to obtain multiple types of target monitoring signals, including: Identify the target influencing factors corresponding to each type of monitoring signal; Based on the target influencing factors corresponding to each type of monitoring signal, a signal compensation model is matched from the preset compensation model library or a signal compensation model corresponding to that type of monitoring signal is constructed based on the target influencing factors corresponding to each type of monitoring signal. The monitoring signals of each type are compensated based on the signal compensation model corresponding to each type of monitoring signal to obtain multi-type target monitoring signals.
[0025] In this embodiment, the preset compensation model library is managed in a categorized and hierarchical manner. Specifically, it is classified in multiple dimensions according to the type of monitoring signal, the combination characteristics of influencing factors, and the system operation stage. Detailed applicable scenario descriptions, performance index parameters, and case verification data are added to each model. Among them, the monitoring signal types include temperature, pressure, displacement, etc.; the system operation stages include initial debugging, normal operation, aging stage, etc.
[0026] In this embodiment, a compensation model is preferentially matched from a preset compensation model library; if no compensation model is matched, a signal compensation model is constructed based on the target influencing factors corresponding to each type of monitoring signal; specifically, this includes: By using physical mechanism analysis, a basic mathematical model is established between the target influencing factors and the monitoring signals. Combining machine learning algorithms (random forest, etc.), the first monitoring signal set (original monitoring signal data) and the second monitoring signal set (monitoring signal data obtained under ideal conditions) are used as training samples to optimize the parameters and adjust the structure of the basic mathematical model, thus obtaining the signal compensation model.
[0027] In addition, during model training, methods such as cross-validation and leave-one-out method are used to evaluate and optimize the model to avoid overfitting and ensure that the constructed signal compensation model has good generalization ability and accuracy.
[0028] In this embodiment, after compensating the monitoring signal based on the signal compensation model corresponding to each type of monitoring signal, the quality of the compensated target monitoring signal is evaluated using methods such as statistical analysis and spectrum analysis; the error index between the compensated signal and the ideal signal is calculated; if the error index exceeds the preset threshold, the determination of the target influencing factors, the selection or construction process of the signal compensation model are re-examined, and the model is corrected and optimized until the compensated target monitoring signal meets the accuracy requirements.
[0029] This embodiment makes the monitoring signal compensation process more adaptable and accurate by determining the corresponding influencing factors and then matching or constructing a signal compensation model from a preset compensation model library; it ensures that each type of monitoring signal can be optimally processed, further improving the quality of monitoring data.
[0030] In one possible embodiment, target influencing factors corresponding to each type of monitoring signal are determined. For example, target influencing factors corresponding to temperature signals include ambient temperature fluctuations, target influencing factors corresponding to pressure signals include changes in the self-weight stress of soil and rock, target influencing factors corresponding to displacement signals include seasonal freeze-thaw cycles, and target influencing factors corresponding to humidity signals include groundwater seepage. Based on the target influencing factors corresponding to each type of monitoring signal, a signal compensation model is matched from a preset compensation model library or a signal compensation model corresponding to that type of monitoring signal is constructed based on the target influencing factors. The preset compensation model library pre-stores temperature compensation models based on linear regression algorithms, pressure compensation models based on finite element analysis methods, displacement compensation models based on time series analysis algorithms, and humidity compensation models based on Darcy's law. If a compensation model matching the target influencing factor exists in the preset compensation model library, it is directly called; otherwise, based on the relevant data of the target influencing factors, a corresponding signal compensation model is constructed based on the target influencing factors corresponding to each type of monitoring signal. The signal compensation model corresponding to each type of monitoring signal is used to compensate for that type of monitoring signal to obtain multiple types of target monitoring signals.
[0031] In one embodiment of this application, a signal compensation model is constructed based on the target influencing factors corresponding to each type of monitoring signal, including: The compensation factor corresponding to each type of monitoring signal is determined based on the first set of monitoring signals corresponding to each type of monitoring signal and the second set of monitoring signals corresponding to that type of monitoring signal. The first model is trained based on the first monitoring signal set corresponding to each type of monitoring signal and the compensation factor corresponding to that type of monitoring signal to obtain the signal compensation model corresponding to that type of monitoring signal. The first monitoring signal set consists of raw monitoring signal data; the second monitoring signal set consists of monitoring signal data acquired under ideal conditions.
[0032] In this embodiment, if there is no compensation model matching the target influencing factor in the preset compensation model library, the compensation factor corresponding to the monitoring signal is determined according to the first monitoring signal set corresponding to each type of monitoring signal and the second monitoring signal set corresponding to the monitoring signal.
[0033] Specifically, the second monitoring signal set consists of monitoring signal data acquired under controlled experimental conditions or calibrated ideal operating conditions. The ideal state is defined as: ambient temperature fluctuations less than ±0.5℃, stable groundwater levels, no external construction vibrations, the soil and rock mass in a stable stress state, and sensors calibrated with standard instruments and measurement errors less than 1%FS. The methods for acquiring the second monitoring signal set include: in a laboratory soil and rock energy storage simulation device, installing temperature, pressure, displacement, and humidity sensors on a standard sample, controlling environmental conditions to meet the aforementioned ideal state, collecting data for at least 24 hours, and taking the average value as the ideal signal; for field systems, data can be screened during stable operating periods after shutdown and maintenance, or the second monitoring signal set can be obtained by synchronous measurement using calibrated standard instruments.
[0034] For each type of monitoring signal, the first monitoring signal set (raw monitoring signal data) and the second monitoring signal set (monitoring signal data obtained under ideal conditions) are compared point by point, the average difference between the two is calculated, and then weighted by combining the historical relevant influencing factor data to obtain the compensation factor corresponding to that type of monitoring signal.
[0035] Alternatively, the first and second monitoring signal sets can be spatiotemporally aligned to construct a time series comparison matrix. Multi-scale feature analysis can be used to extract feature components of different frequency bands, and sensitivity analysis can be performed in conjunction with the physical characteristics of the target influencing factors to determine the compensation factor corresponding to this type of monitoring signal.
[0036] In this embodiment, a first model is trained based on a first monitoring signal set corresponding to each type of monitoring signal and a compensation factor corresponding to that type of monitoring signal to obtain a signal compensation model corresponding to that type of monitoring signal. The first monitoring signal set is used as the input sample, a second monitoring signal set is used as the supervision target, and the compensation factor is used as a constraint or weighting term to construct the loss function of the first model. The parameters of the first model are iteratively updated by minimizing the loss function to obtain the signal compensation model corresponding to that type of monitoring signal. The first monitoring signal set is the original monitoring signal data; the second monitoring signal set is the monitoring signal data obtained under ideal conditions; and the signal compensation model is used to output the compensated signal or compensation amount for that type of monitoring signal. The first model employs a linear regression model, using the first monitoring signal set as input data and the compensation factor as a constraint. The optimization objective is the mean square error between the first model's output on the first monitoring signal set and the second monitoring signal set. The first monitoring signal set is input into the first model to obtain its output. The mean square error between the first model's output and the second monitoring signal set is used as the main loss term, and the deviation of the compensation factor is used as a regularization term to construct the total loss function. The parameters of the first model are iteratively updated using a gradient descent algorithm. When the mean square error converges to a preset accuracy threshold, training is complete, resulting in the signal compensation model corresponding to this type of monitoring signal.
[0037] In this embodiment, the first model is trained based on the first monitoring signal set corresponding to each type of monitoring signal and the compensation factor corresponding to that type of monitoring signal to obtain the signal compensation model corresponding to that type of monitoring signal. The basic framework of the first model can also be constructed using a deep learning algorithm, with the first monitoring signal set as the input sequence and the corresponding compensation factor as the supervision signal, and the parameters are trained using the gradient descent algorithm. In this embodiment, an attention mechanism can also be introduced during the training process to identify the importance weights of the target influencing factors under different time scales and working conditions, thereby enhancing the model's adaptability to complex environments.
[0038] In this embodiment, a signal compensation model is constructed by determining compensation factors based on original monitoring signal data and ideal state data and training the signal compensation model. This method closely combines actual operating data with ideal state requirements and can accurately reflect the relationship between target influencing factors and monitoring signals. It can stably and accurately compensate monitoring signals under different operating conditions and environments, thereby improving the accuracy and effectiveness of the entire monitoring process.
[0039] In one embodiment of this application, determining the target state of a target geotechnical energy storage system based on predicted state data and a preset state feature database includes: Feature extraction is performed on the predicted state data to obtain the target state feature data; Calculate the similarity between the target state feature data and each state feature data pre-stored in the state feature database to obtain multiple similarity values; Based on multiple similarity values, a target similarity value is determined, and the target state of the target geotechnical energy storage system is determined by matching the target similarity value from the state feature database.
[0040] In this embodiment, a multi-scale decomposition technique is used to decompose the predicted state data in multiple dimensions such as time domain, frequency domain, and time-frequency domain to obtain feature components at different scales; principal component analysis is used to eliminate redundant information between feature components and extract the most representative principal components as target state feature data.
[0041] In this embodiment, the similarity between the target state feature data and the pre-stored state feature data in the state feature database is calculated using cosine similarity, Euclidean distance, or dynamic time warping algorithm, resulting in multiple similarity values. Alternatively, this embodiment can employ a kernel-based similarity measurement method to more accurately calculate the similarity between the target state feature data and the pre-stored state feature data in the state feature database, yielding multiple similarity values. In this embodiment, a fuzzy comprehensive evaluation model can be constructed and combined with expert experience and historical data to assign weights to different types of similarity values, and the target similarity value can be calculated comprehensively. Based on the data distribution, model parameters and other information in the similarity calculation process, the confidence level of the target similarity value is calculated. When the target similarity value is higher than a preset threshold and the confidence level meets the requirements, the target state of the target geotechnical energy storage system is directly determined by matching from the state feature database. If the conditions are not met, a clustering analysis algorithm is further used to cluster similar state feature data, and the target state of the system is inferred from the clustering results by combining the system's historical operating data and current operating conditions.
[0042] In this embodiment, by extracting features from the predicted state data and calculating similarity and matching methods, the target state of the system is determined from a preset state feature database, thereby realizing the quantitative analysis and matching of the system's operating state and effectively reducing the probability of state misjudgment.
[0043] In a specific embodiment, determining the target state of the target geotechnical energy storage system based on predicted state data and a preset state feature database may further include: using principal component analysis (PCA) to reduce the dimensionality of the predicted state data, extracting principal components that reflect the main features of the data, and obtaining target state feature data; calculating the similarity between the target state feature data and each state feature data pre-stored in the state feature database, and using a cosine similarity algorithm to calculate the cosine of the angle between the two to obtain multiple similarity values; determining a target similarity value based on the multiple similarity values, selecting the maximum value among the multiple similarity values as the target similarity value, and matching and determining the target state of the target geotechnical energy storage system from the state feature database according to the target similarity value; if the target similarity value is greater than or equal to a preset similarity threshold, then the state category corresponding to the target similarity value is determined as the target state; if the target similarity value is less than the preset similarity threshold, then the target geotechnical energy storage system is determined to be in an unknown state, and an anomaly warning mechanism is triggered. In one embodiment of this application, the monitoring method for a geotechnical energy storage system further includes: The health index of the target geotechnical energy storage system is calculated based on the target state. Based on the health index, determine the monitoring strategy for the target geotechnical energy storage system; The target geotechnical energy storage system was monitored based on the monitoring strategy, and the monitoring results were obtained.
[0044] In this embodiment, the health index of the target geotechnical energy storage system is calculated based on the target state. Specifically, a health index calculation model is established, the target state is mapped to different scoring intervals, and the health index is calculated by weighted average method based on the distribution weight of the target state in each scoring interval. The normal operation state corresponds to the high scoring interval, the early warning state corresponds to the medium scoring interval, the fault state corresponds to the low scoring interval, and the unknown state corresponds to the lowest scoring interval. In this embodiment, the monitoring strategy for the target geotechnical energy storage system is determined based on the health index. Specifically, it includes: when the health index is greater than or equal to a preset high health threshold, a conventional monitoring strategy is adopted, i.e., maintaining the sampling frequency and monitoring cycle of the existing sensors; when the health index is less than the preset high health threshold but greater than or equal to a preset medium health threshold, an enhanced monitoring strategy is adopted, increasing the sampling frequency of the sensors to 2 times / minute and shortening the monitoring cycle; when the health index is less than the preset medium health threshold, an intensive monitoring strategy is adopted, activating backup sensors, increasing the sampling frequency of all sensors to 5 times / minute, and simultaneously uploading monitoring data in real time. In this embodiment, the target soil and rock energy storage system is monitored based on a monitoring strategy to obtain monitoring results. Specifically, the operating parameters of various sensors are controlled according to the determined monitoring strategy, and the collected monitoring data is processed and analyzed in real time to output a monitoring result report containing the trend of soil and rock state changes, abnormal data identification, etc. In this embodiment, the calculation of the health index further includes: constructing a multi-level health assessment index system, dividing key parameters in the target state, such as the degree of temperature anomaly, pressure fluctuation range, and displacement change rate, into different levels of indicators according to their importance to system performance. This embodiment uses the analytic hierarchy process (AHP) and the entropy weight method, and determines the weights of each level of indicators based on expert experience and objective data; through a fuzzy comprehensive evaluation model, the target state data is mapped to a fuzzy set of health levels, and the membership degree values of each health level are calculated using a membership function, and then weighted summation is performed to obtain the health index of the target geotechnical energy storage system.
[0045] In this embodiment, multiple health level intervals are divided based on the health index, and differentiated basic monitoring strategies are formulated for each interval. For example, when the health index is in the good interval, the frequency of routine monitoring signal acquisition is reduced to decrease data transmission and storage pressure; when the health index enters the warning interval, the monitoring frequency of key parameters is increased, and the data acquisition time interval is encrypted. Combining the system structure characteristics and operating conditions, spatial analysis algorithms are used to identify high-risk areas as key monitoring areas. Distributed sensor arrays are used in key monitoring areas to achieve high-density, multi-dimensional monitoring. In addition, an adaptive adjustment mechanism for the monitoring strategy is established. When the system operating state changes abruptly or external environmental factors have a significant impact, reinforcement learning algorithms are used to dynamically optimize the monitoring signal acquisition frequency, monitoring equipment layout, and data transmission path based on real-time data, forming a dynamic monitoring strategy.
[0046] In this embodiment, a health index is calculated based on the target state, and a monitoring strategy is determined based on the health index. This enables dynamic and intelligent allocation of monitoring resources. For systems with a good health index, the monitoring frequency is appropriately reduced to decrease manpower and material costs. For systems with a low health index, the monitoring intensity is increased to promptly capture potential risks and effectively improve monitoring efficiency.
[0047] In one embodiment of this application, determining a monitoring strategy for a target geotechnical energy storage system based on a health index includes: The basic acquisition frequency of the corresponding monitoring signal is determined based on the preset health index range in which the health index falls. Key monitoring areas are identified based on health indices and / or target status. Among them, the collection frequency of key monitoring areas is higher than that of non-key areas.
[0048] In this embodiment, the monitoring strategy for the target geotechnical energy storage system is determined based on the health index, specifically including: determining the corresponding basic acquisition frequency of monitoring signals according to the preset health index range in which the health index falls. Preferably, the high health threshold is 80 points and the medium health threshold is 60 points. When the health index is greater than or equal to the high health threshold, the basic collection frequency is the normal frequency, for example, 1 time / minute. When the medium health threshold (e.g., 60 points) is less than or equal to the health index and the high health threshold, the basic collection frequency is increased to the enhanced frequency, for example, 2 times / minute. When the health index is less than the medium health threshold, the basic collection frequency is increased to the intensive frequency, for example, 5 times / minute.
[0049] In this embodiment, if the health index is less than the medium health threshold, or the target state is a warning state, a fault state, or an unknown state, the key monitoring area identification process is triggered: Based on historical fault data and geotechnical mechanics models, geotechnical areas with high correlation to the current target state are located, such as areas with abnormally high temperatures and areas with sudden displacement changes, and marked as key monitoring areas. For sensors in key monitoring areas, such as temperature / pressure sensors at corresponding locations, priority acquisition frequencies are set, for example, increasing them by 50% on top of the basic acquisition frequency. Non-critical areas maintain the basic acquisition frequency to reduce data transmission bandwidth usage.
[0050] This application achieves three levels of monitoring intensity by mapping different base frequencies across health index ranges, balancing monitoring accuracy and system power consumption. By combining health index with target status, high-risk areas are located, and the efficiency of abnormal data capture is improved by prioritizing collection frequencies, avoiding the waste of resources from high-frequency collection across the entire area.
[0051] In one embodiment of this application, the monitoring method for a geotechnical energy storage system further includes: Obtain the environmental impact characteristics of the target geotechnical energy storage system; The environmental impact characteristics meet the environmental impact conditions. Based on the environmental impact characteristics, the basic acquisition frequency of the monitoring signal is adjusted to obtain the target monitoring signal acquisition frequency, and the range of the key monitoring area is expanded based on the environmental impact characteristics. Based on the target monitoring signal acquisition frequency and the expanded key monitoring area, a second monitoring strategy for the target geotechnical energy storage system is determined. The target geotechnical energy storage system was monitored based on the second monitoring strategy, and the second monitoring results were obtained.
[0052] In this embodiment, environmental impact characteristics of the target geotechnical energy storage system are acquired in real time through multiple channels, including meteorological monitoring stations, geological monitoring networks, and sensors in surrounding facilities. These characteristics include, but are not limited to, extreme weather data (meteorological parameters such as heavy rain, heavy snow, high temperatures, and strong winds), geological activity data (seismic wave signals, groundwater level changes, and soil moisture fluctuations), and surrounding engineering activity data (vibration frequencies from nearby construction sites and information on underground pipeline modifications). Outliers are removed using data cleaning algorithms, data format is unified using normalization methods, and spatiotemporal interpolation techniques are combined to transform discrete environmental data into continuous feature vectors, thus constructing a multidimensional environmental impact feature dataset.
[0053] In this embodiment, an environmental impact condition judgment model based on the fusion of rule engine and machine learning is constructed. On the one hand, basic judgment rules are preset, such as triggering a preliminary warning when rainfall exceeds 50 mm / hour, earthquake magnitude reaches level 3 or above, or the vibration frequency of surrounding construction exceeds 20 Hz. On the other hand, a support vector machine model is trained using historical environmental data and system response data to learn the impact patterns of different combinations of environmental factors on the geotechnical energy storage system, thereby achieving intelligent classification of environmental impact conditions. The degree of environmental impact is divided into three levels: mild, moderate, and severe, with different monitoring strategies adjusting thresholds and priorities for each level.
[0054] In this embodiment, when the environmental impact characteristics meet the corresponding level of environmental impact conditions, a differentiated adjustment strategy is adopted based on the type and intensity of the environmental impact characteristics. In cases where extreme weather (such as heavy rain) causes a rapid rise in groundwater levels, the basic acquisition frequency of groundwater-related monitoring signals is increased to 3-5 times the original frequency. Simultaneously, an area within a 50-meter radius of the area with significant water level changes is included in the expanded key monitoring area. Regarding the impact of vibration from surrounding construction, the key monitoring area is expanded in the direction of the vibration source, and accelerometers are deployed to increase the vibration monitoring signal acquisition frequency to 100Hz. In a specific embodiment, the basic acquisition frequency of monitoring signals is adjusted based on environmental impact characteristics to obtain the target monitoring signal acquisition frequency. Specifically, when the regional rainfall exceeds a preset threshold, the basic acquisition frequency is increased by one level; when the ambient temperature changes drastically beyond a preset range, the acquisition frequency of temperature-related monitoring signals is increased to the highest level; when the intensity of seismic activity reaches a certain level, the acquisition frequency of all monitoring signals is increased to the highest level. Based on environmental impact characteristics, the scope of key monitoring areas is expanded as follows: when regional rainfall exceeds a preset threshold, the soil and rock areas near drainage channels and groundwater level monitoring points are expanded into key monitoring areas; when the ambient temperature changes drastically beyond a preset range, areas with frequent heat exchange and a certain distance around them are expanded into key monitoring areas; when the intensity of seismic activity reaches a certain level, the weak structural areas of the soil and rock energy storage system and areas with dense sensor installations are expanded into key monitoring areas.
[0055] The second monitoring strategy is obtained by updating the basic collection frequency and key monitoring areas in the first monitoring strategy when the environmental impact characteristics meet the environmental impact conditions; when the environmental impact conditions are lifted, the first monitoring strategy can be restored or the monitoring strategy can be redefined.
[0056] A monitoring method for a geotechnical energy storage system corresponding to the above embodiment, Figure 2 This is a structural block diagram of a monitoring device for a geotechnical energy storage system according to an embodiment of this application. For ease of explanation, only the parts relevant to the embodiment of this application are shown. References Figure 2 The monitoring device 20 of the geotechnical energy storage system includes: a signal acquisition module 21, a signal compensation module 22, a data processing module 23, and a status determination module 24.
[0057] Among them, the signal acquisition module 21 is used to acquire multiple monitoring signals of the target geotechnical energy storage system; Signal compensation module 22 is used to compensate for each type of monitoring signal based on the target influencing factors to obtain multiple types of target monitoring signals; Among them, the target influencing factors corresponding to each type of monitoring signal are different. The target influencing factors are selected from the set of preset influencing factors that affect the performance of the target geotechnical energy storage system. Data processing module 23 is used to input multiple types of target monitoring signals into the state prediction model for processing, and obtain the predicted state data of the target geotechnical energy storage system; The state determination module 24 is used to determine the target state of the target geotechnical energy storage system based on the predicted state data and the preset state feature database.
[0058] In one embodiment of this application, the signal compensation module 22 is specifically used for: Identify the target influencing factors corresponding to each type of monitoring signal; Based on the target influencing factors corresponding to each type of monitoring signal, a signal compensation model is matched from the preset compensation model library or a signal compensation model corresponding to that type of monitoring signal is constructed based on the target influencing factors corresponding to each type of monitoring signal. The monitoring signals of each type are compensated based on the signal compensation model corresponding to each type of monitoring signal to obtain multi-type target monitoring signals.
[0059] In one embodiment of this application, the signal compensation module 22 is specifically used for: The compensation factor corresponding to each type of monitoring signal is determined based on the first set of monitoring signals corresponding to each type of monitoring signal and the second set of monitoring signals corresponding to that type of monitoring signal. The first model is trained based on the first monitoring signal set corresponding to each type of monitoring signal and the compensation factor corresponding to that type of monitoring signal to obtain the signal compensation model corresponding to that type of monitoring signal. The first monitoring signal set consists of raw monitoring signal data; the second monitoring signal set consists of monitoring signal data acquired under ideal conditions.
[0060] In one embodiment of this application, the state determination module 24 is specifically used for: Feature extraction is performed on the predicted state data to obtain the target state feature data; Calculate the similarity between the target state feature data and each state feature data pre-stored in the state feature database to obtain multiple similarity values; Based on multiple similarity values, a target similarity value is determined, and the target state of the target geotechnical energy storage system is determined by matching the target similarity value from the state feature database.
[0061] In one embodiment of this application, the monitoring device 20 for the geotechnical energy storage system further includes: a monitoring module, specifically used for: The health index of the target geotechnical energy storage system is calculated based on the target state. Based on the health index, determine the monitoring strategy for the target geotechnical energy storage system; The target geotechnical energy storage system was monitored based on the monitoring strategy, and the monitoring results were obtained.
[0062] In one embodiment of this application, the monitoring module is specifically used for: The basic acquisition frequency of the corresponding monitoring signal is determined based on the preset health index range in which the health index falls. Key monitoring areas are identified based on health indices and / or target status. Among them, the collection frequency of key monitoring areas is higher than that of non-key areas.
[0063] In one embodiment of this application, the monitoring module is specifically used for: Obtain the environmental impact characteristics of the target geotechnical energy storage system; The environmental impact characteristics meet the environmental impact conditions. Based on the environmental impact characteristics, the basic acquisition frequency of the monitoring signal is adjusted to obtain the target monitoring signal acquisition frequency, and the range of the key monitoring area is expanded based on the environmental impact characteristics. Based on the target monitoring signal acquisition frequency and the expanded key monitoring area, a second monitoring strategy for the target geotechnical energy storage system is determined. The target geotechnical energy storage system was monitored based on the second monitoring strategy, and the second monitoring results were obtained.
[0064] See Figure 3 , Figure 3 This is a schematic block diagram of an electronic device provided according to an embodiment of this application. Figure 3 The electronic device 300 in this embodiment may include one or more processors 301, one or more input devices 302, one or more output devices 303, and one or more memories 304. The processors 301, input devices 302, output devices 303, and memories 304 communicate with each other via a communication bus 305. The memories 304 store computer programs, including program instructions. The processors 301 execute the program instructions stored in the memories 304. Specifically, the processors 301 are configured to invoke the program instructions to perform the functions of the modules in the aforementioned device embodiments, for example... Figure 2 The functions of the signal acquisition module 21, signal compensation module 22, data processing module 23, and status determination module 24 are shown.
[0065] It should be understood that, in the embodiments of this application, the processor 301 may be a central processing unit (CPU), or it may be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor.
[0066] Input device 302 may include a touchpad, a fingerprint sensor (for collecting the user's fingerprint information and fingerprint orientation information), a microphone, etc., and output device 303 may include a display (LCD, etc.), a speaker, etc.
[0067] The memory 304 may include read-only memory and random access memory, and provides instructions and data to the processor 301. A portion of the memory 304 may also include non-volatile random access memory. For example, the memory 304 may also store device type information.
[0068] In specific implementations, the processor 301, input device 302, and output device 303 described in the embodiments of this application can execute the implementation methods described in any embodiment of the monitoring method for a geotechnical energy storage system provided in the embodiments of this application, or they can execute the implementation methods of the electronic devices described in the embodiments of this application, which will not be repeated here.
[0069] In another embodiment of this application, a computer-readable storage medium is provided. This computer-readable storage medium stores a computer program, which includes program instructions. When executed by a processor, the program instructions implement all or part of the processes in the methods described above. Alternatively, the computer program can instruct related hardware to complete the process. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include any entity or device capable of carrying computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.
[0070] The computer-readable storage medium can be an internal storage unit of the electronic device in any of the foregoing embodiments, such as a hard disk or memory of the electronic device. The computer-readable storage medium can also be an external storage device of the electronic device, such as a plug-in hard disk, smart media card (SMC), secure digital card (SD), flash card, etc., equipped on the electronic device. Furthermore, the computer-readable storage medium can include both internal and external storage units of the electronic device. The computer-readable storage medium is used to store computer programs and other programs and data required by the electronic device. The computer-readable storage medium can also be used to temporarily store data that has been output or will be output.
[0071] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this application.
[0072] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the electronic devices and units described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.
[0073] In the several embodiments provided in this application, it should be understood that the disclosed electronic devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces or units, or it may be an electrical, mechanical, or other form of connection.
[0074] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of the embodiments of this application, depending on actual needs.
[0075] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.
[0076] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.
Claims
1. A monitoring method for a geotechnical energy storage system, characterized in that, include: Acquire various monitoring signals from the target geotechnical energy storage system; Each type of monitoring signal is compensated based on the target influencing factors to obtain multiple types of target monitoring signals; Among them, each type of monitoring signal corresponds to a different target influencing factor, which is selected from a set of preset influencing factors that affect the performance of the target geotechnical energy storage system; The multi-target monitoring signals are input into the state prediction model for processing to obtain the predicted state data of the target geotechnical energy storage system. The target state of the target geotechnical energy storage system is determined based on the predicted state data and the preset state feature database.
2. The monitoring method for a geotechnical energy storage system according to claim 1, characterized in that, The compensation for each type of monitoring signal based on the target influencing factors yields multiple types of target monitoring signals, including: Identify the target influencing factors corresponding to each type of monitoring signal; Based on the target influencing factors corresponding to each type of monitoring signal, a signal compensation model is matched from the preset compensation model library or a signal compensation model corresponding to that type of monitoring signal is constructed based on the target influencing factors corresponding to each type of monitoring signal. The monitoring signals of each type are compensated based on the signal compensation model corresponding to each type of monitoring signal to obtain the multi-type target monitoring signals.
3. The monitoring method for a geotechnical energy storage system according to claim 2, characterized in that, The construction of a signal compensation model for each type of monitoring signal based on the target influencing factors includes: The compensation factor corresponding to each type of monitoring signal is determined based on the first set of monitoring signals corresponding to each type of monitoring signal and the second set of monitoring signals corresponding to that type of monitoring signal. The first model is trained based on the first monitoring signal set corresponding to each type of monitoring signal and the compensation factor corresponding to that type of monitoring signal to obtain the signal compensation model corresponding to that type of monitoring signal. Wherein, the first monitoring signal set is the original monitoring signal data; the second monitoring signal set is the monitoring signal data obtained under ideal conditions.
4. The monitoring method for a geotechnical energy storage system according to claim 1, characterized in that, Determining the target state of the target geotechnical energy storage system based on the predicted state data and a preset state feature database includes: Feature extraction is performed on the predicted state data to obtain the target state feature data; Calculate the similarity between the target state feature data and each state feature data pre-stored in the state feature database to obtain multiple similarity values; Based on the multiple similarity values, a target similarity value is determined, and the target state of the target geotechnical energy storage system is determined by matching the target similarity value from the state feature database.
5. The monitoring method for a geotechnical energy storage system according to claim 1, characterized in that, Also includes: The health index of the target geotechnical energy storage system is calculated based on the target state. Based on the health index, a monitoring strategy for the target geotechnical energy storage system is determined. The target geotechnical energy storage system is monitored based on the aforementioned monitoring strategy, and the monitoring results are obtained.
6. The monitoring method for a geotechnical energy storage system according to claim 5, characterized in that, The monitoring strategy for the target geotechnical energy storage system based on the health index includes: The basic acquisition frequency of the corresponding monitoring signal is determined based on the preset health index range in which the health index falls. Based on the health index and / or the target status, key monitoring areas are determined; The key monitoring areas are sampled more frequently than the non-key areas.
7. The monitoring method for a geotechnical energy storage system according to claim 6, characterized in that, Also includes: Obtain the environmental impact characteristics of the target geotechnical energy storage system; In response to the environmental impact characteristics satisfying the environmental impact conditions. Based on the environmental impact characteristics, the basic acquisition frequency of the monitoring signal is adjusted to obtain the target monitoring signal acquisition frequency, and the range of the key monitoring area is expanded based on the environmental impact characteristics. Based on the target monitoring signal acquisition frequency and the expanded key monitoring area, a second monitoring strategy for the target geotechnical energy storage system is determined. The target geotechnical energy storage system is monitored based on the second monitoring strategy to obtain the second monitoring result.
8. A monitoring device for a geotechnical energy storage system, characterized in that, include: The signal acquisition module is used to acquire various monitoring signals of the target geotechnical energy storage system; The signal compensation module is used to compensate each type of monitoring signal based on the target influencing factors to obtain multiple types of target monitoring signals; wherein, the target influencing factors corresponding to each type of monitoring signal are different, and the target influencing factors are selected from a preset set of influencing factors that affect the performance of the target geotechnical energy storage system; The data processing module is used to input the multi-type target monitoring signals into the state prediction model for processing, so as to obtain the predicted state data of the target geotechnical energy storage system. The state determination module is used to determine the target state of the target geotechnical energy storage system based on the predicted state data and a preset state feature database.
9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method as described in any one of claims 1 to 7.
10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method as described in any one of claims 1 to 7.
