Battery thermal runaway early warning method, device and equipment of energy storage power station and medium

Abstract

The invention discloses a battery thermal runaway early warning method, device and equipment for an energy storage power station and a medium, the battery thermal runaway risk of the energy storage power station is predicted by using a neural network, early warning is performed before battery thermal runaway of the energy storage power station, and the safety is improved. Besides, the probabilistic neural network is trained by using a small number of training samples, and after the probabilistic neural network is converged, the back-propagation neural network is trained by using the soft label output by the probabilistic neural network, so that the problems of low learning speed and insufficient training samples in the initial training stage of the back-propagation neural network are solved, and the model training efficiency is improved. Meanwhile, the soft label can solve the problem that the back propagation neural network is prone to falling into a local minimum value in the training process.

Description

Technical Field

[0001] The present invention relates to a battery thermal runaway early warning technology, and in particular to a battery thermal runaway early warning method, device, equipment and medium for an energy storage power station. Background Art

[0002] An energy storage power station is a device system that uses electrochemical batteries as energy storage media to store, convert and release cyclical electrical energy, playing the role of balancing the supply and demand of electricity, that is, storing electrical energy during peak load periods of the power system and releasing electrical energy when the load is lower, thereby balancing the difference between supply and demand of electricity.

[0003] Energy storage power stations mostly use lithium-ion batteries as energy storage media, and lithium-ion batteries have a greater risk of thermal runaway during use. Existing research shows that before thermal runaway, the chemical reaction of lithium-ion batteries is relatively slow, but it will release trace gases, smoke and heat. As the internal heat continues to accumulate, when the critical temperature is reached, the battery will experience thermal runaway, triggering a series of violent chain reactions, and ultimately leading to violent combustion or even explosion.

[0004] At present, thermal runaway detection of lithium-ion batteries is mostly based on data collected by sensors such as gas composition detectors, temperature sensors, and photoelectric smoke detectors. Although they can detect drastic changes when thermal runaway occurs, they cannot issue an alarm at the best time before thermal runaway occurs, and there are a lot of false alarms caused by environmental factors. This detection lag and false alarms seriously limit the early warning capabilities of energy storage power stations in the early stages of fires, making it difficult to meet actual application needs. Summary of the invention

[0005] The present invention provides a battery thermal runaway early warning method, device, equipment and medium for an energy storage power station, so as to achieve early warning before the battery thermal runaway of the energy storage power station, thereby improving safety.

[0006] In a first aspect, the present invention provides a battery thermal runaway early warning method for an energy storage power station, comprising:

[0007] Training a probabilistic neural network and a back propagation neural network using a training set until the probabilistic neural network converges, the training set comprising a plurality of first training samples with hard labels, the training samples comprising sensor data collected by at least one sensor, the hard label indicating whether a battery of the energy storage power station has thermal runaway or has not thermal runaway;

[0008] Using the converged probabilistic neural network to predict the second unlabeled training sample to obtain a soft label of the second training sample, wherein the soft label represents a probability value of thermal runaway of the battery of the energy storage power station;

[0009] Training the back propagation neural network using a plurality of second training samples with soft labels until the back propagation neural network converges;

[0010] The converged back propagation neural network is used to process the sensor data collected in real time from the energy storage power station to predict whether there is a risk of thermal runaway of the battery of the energy storage power station;

[0011] When there is a risk of thermal runaway of the battery in the energy storage power station, a thermal runaway warning is issued.

[0012] Optionally, after the sensor data is collected, the following steps are also included:

[0013] Taking the three spatial dimensions of the energy storage power station as three channels, reconstructing the sensor data into three-channel data;

[0014] Inputting the three-channel data into a depth-wise separable convolutional network for processing to obtain a feature map;

[0015] Performing global pooling processing on the feature map to obtain pooled features;

[0016] The pooled features are batch normalized to obtain input features of the probabilistic neural network and the back propagation neural network.

[0017] Optionally, the three-channel data is input into a depthwise separable convolutional network for processing to obtain a feature map, including:

[0018] Performing spatial convolution processing on each channel data of the three-channel data to obtain a spatial convolution feature map;

[0019] The spatial convolution feature maps of each channel data are spliced ​​in dimension to obtain a spliced ​​feature map;

[0020] Perform point-by-point convolution processing on the concatenated feature map to obtain a feature map.

[0021] Optionally, batch normalization is performed on the pooled features to obtain input features of a probabilistic neural network and a back propagation neural network, including:

[0022] Calculating the mean and variance of the pooled features;

[0023] Normalizing the pooled features based on the mean and the variance to obtain normalized features;

[0024] The normalized features are translated and scaled to obtain input features of the probabilistic neural network and the back propagation neural network.

[0025] Optionally, after the converged back propagation neural network is used to process the sensor data collected in real time from the energy storage power station to predict whether the battery of the energy storage power station is at risk of thermal runaway, the method further includes:

[0026] The prediction results are used as training sets to train the probabilistic neural network and the back-propagation neural network, and the network parameters of the probabilistic neural network and the back-propagation neural network are optimized.

[0027] Optionally, the sensing data includes gas composition information measured by a gas composition detector, temperature information measured by a temperature sensor, and visible light information measured by a photoelectric smoke detector.

[0028] In a second aspect, the present invention further provides a battery thermal runaway warning device for an energy storage power station, comprising:

[0029] A first training module is used to train a probabilistic neural network and a back propagation neural network using a training set until the probabilistic neural network converges, the training set includes a plurality of first training samples with hard labels, the training samples include sensor data collected by at least one sensor, and the hard label indicates whether a battery of the energy storage power station has thermal runaway or not;

[0030] a soft label generation module, configured to use the converged probabilistic neural network to predict the second training sample without a label, and obtain a soft label of the second training sample, wherein the soft label represents a probability value of thermal runaway of the battery of the energy storage power station;

[0031] A second training module, used for training the back propagation neural network using a plurality of second training samples with soft labels until the back propagation neural network converges;

[0032] A risk prediction module is used to process the sensor data collected in real time from the energy storage power station by using the converged back propagation neural network to predict whether there is a thermal runaway risk in the battery of the energy storage power station;

[0033] The runaway warning module is used to issue a thermal runaway warning when there is a risk of thermal runaway of the battery in the energy storage power station.

[0034] In a third aspect, the present invention further provides an electronic device, comprising:

[0035] one or more processors;

[0036] A storage device for storing one or more programs;

[0037] When the one or more programs are executed by the one or more processors, the one or more processors implement the battery thermal runaway early warning method for the energy storage power station provided in the first aspect of the present invention.

[0038] In a fourth aspect, the present invention further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the battery thermal runaway warning method for an energy storage power station as provided in the first aspect of the present invention.

[0039] In a fifth aspect, the present invention further provides a computer program product, including a computer program, which, when executed by a processor, implements the battery thermal runaway warning method for an energy storage power station as provided in the first aspect of the present invention.

[0040] The battery thermal runaway early warning method of the energy storage power station provided by the present invention uses a training set to train a probabilistic neural network and a back propagation neural network until the probabilistic neural network converges, the training set includes a plurality of first training samples with hard labels, the training samples include sensor data collected by at least one sensor, the hard label indicates that the battery of the energy storage power station has thermal runaway or has not thermal runaway, the converged probabilistic neural network is used to predict the second training sample without a label, and the soft label of the second training sample is obtained, and the soft label indicates the probability value of the battery of the energy storage power station having thermal runaway, and the back propagation neural network is trained using a plurality of second training samples with soft labels until the back propagation neural network converges, and the sensor data collected in real time from the energy storage power station is processed by the back propagation neural network after convergence, and it is predicted whether the battery of the energy storage power station has a thermal runaway risk, and when the battery of the energy storage power station has a thermal runaway risk, a thermal runaway early warning is issued. The present invention uses a neural network to predict the risk of thermal runaway of the battery of the energy storage power station, and gives an early warning before the battery of the energy storage power station has thermal runaway, thereby improving safety. In addition, a small number of training samples are first used to train the probabilistic neural network. After the probabilistic neural network converges, the soft labels output by the probabilistic neural network are used to train the back propagation neural network, which solves the problems of slow learning speed and insufficient training samples in the early stage of back propagation neural network training and improves the model training efficiency. At the same time, soft labels can solve the problem that the back propagation neural network is prone to falling into local minima during training.

[0041] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0043] Figure 1 A flow chart of a battery thermal runaway early warning method for an energy storage power station provided by the present invention;

[0044] Figure 2 A schematic diagram of training a battery thermal runaway warning model for an energy storage power station provided by the present invention;

[0045] Figure 3 A schematic structural diagram of a battery thermal runaway warning device for an energy storage power station provided by the present invention;

[0046] Figure 4 The present invention provides a schematic structural diagram of an electronic device.

[0047] The above drawings have shown clear embodiments of the present application, which will be described in more detail later. These drawings and text descriptions are not intended to limit the scope of the present application in any way, but to illustrate the concept of the present application to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0048] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0049] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0050] Figure 1 This is a flow chart of a battery thermal runaway warning method for an energy storage power station provided by the present invention. This embodiment can be applied to early warning of battery thermal runaway in an energy storage power station. The method can be executed by a battery thermal runaway warning device for an energy storage power station provided by an embodiment of the present invention. The device can be implemented by software and / or hardware, and is usually configured in an electronic device, such as Figure 1As shown, the battery thermal runaway early warning method of the energy storage power station specifically includes the following steps:

[0051] S101. Train the probabilistic neural network and the back propagation neural network using the training set until the probabilistic neural network converges.

[0052] In an embodiment of the present invention, a small amount of training sets are first obtained, and the training sets include multiple first training samples with hard labels. The training samples include sensor data collected by at least one sensor, and the hard label indicates whether the battery of the energy storage power station has thermal runaway or not. Among them, the sensor data may include gas composition information measured by a gas composition detector, temperature information measured by a temperature sensor, and visible light information measured by a photoelectric smoke detector, and these sensors are arranged inside the energy storage power station.

[0053] Figure 2 A training diagram of a battery thermal runaway warning model for an energy storage power station provided by the present invention, such as Figure 2 As shown, in an embodiment of the present invention, a small amount of first training samples are simultaneously input into a probabilistic neural network (PNN) and a back propagation neural network (BP) for training until the probabilistic neural network converges. The probabilistic neural network is structurally divided into four network layers: input layer, sample layer, summation layer, and competitive output layer. Among them, the input layer is used to calculate the distance between the input vector and all training sample vectors. The main function of the sample layer is to perform a weighted sum operation on the input signal and send it to the next layer after an activation function operation, wherein the activation function is a Gaussian function. The summation layer adds the output of the sample layer by class, and the competitive output layer outputs the judgment result, that is, predicting the probability of the first training sample having a battery thermal runaway risk.

[0054] Compared with traditional neural networks, the biggest advantage of probabilistic neural networks is that they do not need to perform multiple calculations to obtain a converged Bayesian optimization solution. Therefore, when debugging and identifying, using probabilistic neural networks can obtain the desired test results with fewer training times. Therefore, probabilistic neural networks can converge before back-propagation neural networks.

[0055] In some embodiments of the present invention, after obtaining the first training sample, the first training sample may be preprocessed and then sent to the probabilistic neural network and the back propagation neural network for training. Exemplarily, the preprocessing process is as follows:

[0056] 1. Taking the three spatial dimensions of the energy storage power station as three channels, the sensor data is reconstructed into three-channel data.

[0057] In the embodiment of the present invention, the three-dimensional space of the energy storage power station is taken as three channels, and the sensor data is reconstructed into three-channel data. Exemplarily, taking one of the three-dimensional coordinate axes of the space as an example, the sensor data on all sampling sections perpendicular to the coordinate axis are determined as the channel data of the channel, so that the sensor data is reconstructed into three-channel data.

[0058] 2. Input the three-channel data into the deep separable convolutional network for processing to obtain the feature map.

[0059] For example, Figure 2 As shown in the figure, the three-channel data is input into the depthwise separable convolution network (DSC) for processing to obtain the feature map. The depthwise separable convolution is an algorithm obtained by improving the standard convolution calculation in the convolutional neural network. It reduces the number of parameters required for convolution calculation by splitting the correlation between the spatial dimension and the channel (depth) dimension, and improves the efficiency of the use of convolution kernel parameters.

[0060] Exemplarily, spatial convolution processing is performed on each channel data of the three-channel data to obtain a spatial convolution feature map, the spatial convolution feature maps of each channel data are spliced ​​in dimension to obtain a spliced ​​feature map, and point-by-point convolution processing is performed on the spliced ​​feature map to obtain a feature map.

[0061] 3. Perform global pooling on the feature map to obtain pooled features.

[0062] like Figure 2 As shown, the feature map output by the deep separable convolutional network is globally pooled to obtain pooled features. In the embodiment of the present invention, the global pooling process may include global average pooling and global maximum pooling, which are not limited in the embodiment of the present invention. The purpose of pooling is to reduce the dimension and strengthen the features. The specific formula is:

[0063]

[0064] Among them, down is a pooling function, for example, a maximum pooling function or an average pooling function, and s is the size of the pooling window.

[0065] 4. Perform batch normalization on the pooled features to obtain the input features of the probabilistic neural network and the back propagation neural network.

[0066] like Figure 2 As shown in FIG. 1 , the pooled features are batch normalized to obtain the input features of the probabilistic neural network and the back propagation neural network.

[0067] Specifically, the mean and variance of the pooled features are calculated, and the pooled features are normalized based on the mean and variance to obtain the normalized features, and the normalized features are translated and scaled to obtain the input features of the probabilistic neural network and the back propagation neural network. The formula for batch normalization is as follows:

[0068]

[0069] Among them, σ 2 is the variance of all elements in the pooled feature, μ is the mean of all elements in the pooled feature, ε is an infinite decimal, is the element in the normalized feature, and γ and β are the linear parameters obtained by training for scaling and translation normalization.

[0070] S102: Use the converged probabilistic neural network to predict the second training sample without a label to obtain a soft label of the second training sample.

[0071] For example, in actual applications, sensor data in an energy storage power station can be collected as a second unlabeled training sample. After the probabilistic neural network converges, the converged probabilistic neural network is used to predict the second unlabeled training sample to obtain a soft label for the second training sample. The soft label represents the probability value of thermal runaway of the battery in the energy storage power station.

[0072] S103: Train the back propagation neural network using a plurality of second training samples with soft labels until the back propagation neural network converges.

[0073] In the embodiment of the present invention, a plurality of second training samples with soft labels are used to train the back propagation neural network until the back propagation neural network converges, thereby solving the problems of slow learning speed and insufficient training samples in the early stage of back propagation neural network training and improving the model training efficiency. At the same time, since the second training samples have soft labels, the problem of easily falling into local minima during the back propagation neural network training process can be solved.

[0074] S104, using the converged back propagation neural network to process the sensor data collected in real time from the energy storage power station, and predict whether the battery of the energy storage power station has a risk of thermal runaway.

[0075] In an embodiment of the present invention, the converged back propagation neural network is used to process the sensor data collected in real time from the energy storage power station to predict whether the battery of the energy storage power station is at risk of thermal runaway. The back propagation neural network is a multi-layer feedforward network trained by error back propagation (abbreviated as error back propagation). Its algorithm is called BP algorithm. Its basic idea is the gradient descent method, which uses gradient search technology to minimize the mean square error between the actual output value and the expected output value of the network. The back propagation neural network has a strong nonlinear mapping ability and a flexible network structure. By modifying the weights of each neuron, the error signal is minimized and the prediction accuracy is improved.

[0076] S105. When there is a risk of thermal runaway of the battery in the energy storage power station, a thermal runaway warning is issued.

[0077] When there is a risk of thermal runaway of the battery in the energy storage power station, a thermal runaway warning is issued in advance, so as to achieve early warning before the battery in the energy storage power station thermal runaway, thereby improving safety.

[0078] In the embodiment of the present invention, Figure 2 As shown, after the converged back propagation neural network is used to process the sensor data collected in real time from the energy storage power station and predict whether the battery of the energy storage power station has a risk of thermal runaway, it also includes:

[0079] The prediction results are used as training sets to train the probabilistic neural network and the back propagation neural network, and the network parameters of the probabilistic neural network and the back propagation neural network are continuously optimized to improve the prediction accuracy of the model.

[0080] The battery thermal runaway early warning method of the energy storage power station provided by the present invention uses a training set to train a probabilistic neural network and a back propagation neural network until the probabilistic neural network converges, the training set includes a plurality of first training samples with hard labels, the training samples include sensor data collected by at least one sensor, the hard label indicates that the battery of the energy storage power station has thermal runaway or has not thermal runaway, the converged probabilistic neural network is used to predict the second training sample without a label, and the soft label of the second training sample is obtained, and the soft label indicates the probability value of the battery of the energy storage power station having thermal runaway, and the back propagation neural network is trained using a plurality of second training samples with soft labels until the back propagation neural network converges, and the sensor data collected in real time from the energy storage power station is processed by the back propagation neural network after convergence, and it is predicted whether the battery of the energy storage power station has a thermal runaway risk, and when the battery of the energy storage power station has a thermal runaway risk, a thermal runaway early warning is issued. The present invention uses a neural network to predict the risk of thermal runaway of the battery of the energy storage power station, and gives an early warning before the battery of the energy storage power station has thermal runaway, thereby improving safety. In addition, a small number of training samples are first used to train the probabilistic neural network. After the probabilistic neural network converges, the soft labels output by the probabilistic neural network are used to train the back propagation neural network, which solves the problems of slow learning speed and insufficient training samples in the early stage of back propagation neural network training and improves the model training efficiency. At the same time, soft labels can solve the problem that the back propagation neural network is prone to falling into local minima during training.

[0081] Figure 3 A schematic diagram of the structure of a battery thermal runaway warning device for an energy storage power station provided by the present invention, such as Figure 3 As shown, the battery thermal runaway warning device of the energy storage power station includes:

[0082] A first training module 201 is used to train a probabilistic neural network and a back propagation neural network using a training set until the probabilistic neural network converges, the training set includes a plurality of first training samples with hard labels, the training samples include sensor data collected by at least one sensor, and the hard label indicates whether a battery of the energy storage power station has thermal runaway or not;

[0083] A soft label generation module 202 is used to use the converged probabilistic neural network to predict the second training sample without a label to obtain a soft label of the second training sample, wherein the soft label represents a probability value of thermal runaway of the battery of the energy storage power station;

[0084] A second training module 203 is used to train the back propagation neural network using a plurality of second training samples with soft labels until the back propagation neural network converges;

[0085] The risk prediction module 204 is used to process the sensor data collected in real time from the energy storage power station by using the converged back propagation neural network to predict whether there is a risk of thermal runaway of the battery in the energy storage power station;

[0086] The runaway warning module 205 is used to issue a thermal runaway warning when there is a risk of thermal runaway of the battery in the energy storage power station.

[0087] In some embodiments of the present invention, the battery thermal runaway warning device of the energy storage power station further includes:

[0088] A data reconstruction module, for reconstructing the sensor data into three-channel data after collecting the sensor data, taking the three spatial dimensions of the energy storage power station as three channels;

[0089] A separation convolution module, used for inputting the three-channel data into a deep separable convolutional network for processing to obtain a feature map;

[0090] A pooling module, used to perform global pooling processing on the feature map to obtain pooling features;

[0091] The normalization module is used to perform batch normalization processing on the pooled features to obtain input features of the probabilistic neural network and the back propagation neural network.

[0092] In some embodiments of the invention, the separation convolution module comprises:

[0093] A spatial convolution submodule, used to perform spatial convolution processing on each channel data of the three-channel data to obtain a spatial convolution feature map;

[0094] A feature splicing submodule, used for splicing the spatial convolution feature maps of each channel data in dimension to obtain a spliced ​​feature map;

[0095] The point-by-point convolution submodule is used to perform point-by-point convolution processing on the concatenated feature map to obtain a feature map.

[0096] In some embodiments of the present invention, the normalization module includes:

[0097] A mean-variance calculation submodule, used to calculate the mean and variance of the pooled feature;

[0098] A normalization submodule, used for normalizing the pooled features based on the mean and the variance to obtain normalized features;

[0099] The translation and scaling submodule is used to translate and scale the normalized features to obtain input features of the probabilistic neural network and the back propagation neural network.

[0100] In some embodiments of the present invention, the battery thermal runaway warning device of the energy storage power station further includes:

[0101] The model optimization module is used to process the sensor data collected in real time from the energy storage power station using the converged back propagation neural network, predict whether the battery of the energy storage power station has a risk of thermal runaway, use the prediction results as a training set, train the probabilistic neural network and the back propagation neural network, and optimize the network parameters of the probabilistic neural network and the back propagation neural network.

[0102] In some embodiments of the present invention, the sensing data includes gas composition information measured by a gas composition detector, temperature information measured by a temperature sensor, and visible light information measured by a photoelectric smoke detector.

[0103] The battery thermal runaway warning device of the energy storage power station can execute the battery thermal runaway warning method of the energy storage power station provided by the aforementioned embodiment of the present invention, and has the corresponding functional modules and beneficial effects of executing the battery thermal runaway warning method of the energy storage power station.

[0104] Figure 4 A schematic diagram of the structure of an electronic device provided by the present invention. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0105] like Figure 4 As shown, the electronic device includes at least one processor 11, and a memory connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 to the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0106] A number of components in the electronic device are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0107] The processor 11 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as a battery thermal runaway early warning method for an energy storage power station.

[0108] In some embodiments, the battery thermal runaway warning method for an energy storage power station may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on an electronic device via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the battery thermal runaway warning method for the energy storage power station described above may be executed. Alternatively, in other embodiments, the processor 11 may be configured to execute the battery thermal runaway warning method for the energy storage power station in any other appropriate manner (e.g., by means of firmware).

[0109] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0110] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0111] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0112] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0113] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0114] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0115] An embodiment of the present invention further provides a computer program product, including a computer program, which, when executed by a processor, implements the battery thermal runaway early warning method for an energy storage power station as provided in any embodiment of the present application.

[0116] In the process of implementation, the computer program product can be written in one or more programming languages ​​or a combination thereof to perform the computer program code of the present invention, including object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (for example, using an Internet service provider to connect through the Internet).

[0117] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0118] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A battery thermal runaway early warning method for an energy storage power station, characterized in that: include: Training a probabilistic neural network and a back propagation neural network using a training set until the probabilistic neural network converges, the training set comprising a plurality of first training samples with hard labels, the training samples comprising sensor data collected by at least one sensor, the hard label indicating whether a battery of the energy storage power station has thermal runaway or has not thermal runaway; Using the converged probabilistic neural network to predict the second unlabeled training sample to obtain a soft label of the second training sample, wherein the soft label represents a probability value of thermal runaway of the battery of the energy storage power station; Training the back propagation neural network using a plurality of second training samples with soft labels until the back propagation neural network converges; The converged back propagation neural network is used to process the sensor data collected in real time from the energy storage power station to predict whether there is a risk of thermal runaway of the battery of the energy storage power station; When there is a risk of thermal runaway of the battery in the energy storage power station, a thermal runaway warning is issued.

2. The battery thermal runaway early warning method for an energy storage power station according to claim 1, characterized in that: After collecting sensor data, it also includes: Taking the three spatial dimensions of the energy storage power station as three channels, reconstructing the sensor data into three-channel data; Inputting the three-channel data into a depth-wise separable convolutional network for processing to obtain a feature map; Performing global pooling processing on the feature map to obtain pooled features; The pooled features are batch normalized to obtain input features of the probabilistic neural network and the back propagation neural network.

3. The battery thermal runaway early warning method for an energy storage power station according to claim 2, characterized in that: The three-channel data is input into a depth-separable convolutional network for processing to obtain a feature map, including: Performing spatial convolution processing on each channel data of the three-channel data to obtain a spatial convolution feature map; The spatial convolution feature maps of each channel data are spliced ​​in dimension to obtain a spliced ​​feature map; Perform point-by-point convolution processing on the concatenated feature map to obtain a feature map.

4. The battery thermal runaway early warning method for an energy storage power station according to claim 2, characterized in that: The pooled features are batch normalized to obtain input features of the probabilistic neural network and the back propagation neural network, including: Calculating the mean and variance of the pooled features; Normalizing the pooled features based on the mean and the variance to obtain normalized features; The normalized features are translated and scaled to obtain input features of the probabilistic neural network and the back propagation neural network.

5. The battery thermal runaway early warning method for an energy storage power station according to any one of claims 1 to 3, characterized in that: After the converged back propagation neural network is used to process the sensor data collected in real time from the energy storage power station to predict whether there is a risk of thermal runaway of the battery of the energy storage power station, the method further includes: The prediction results are used as training sets to train the probabilistic neural network and the back-propagation neural network, and the network parameters of the probabilistic neural network and the back-propagation neural network are optimized.

6. The battery thermal runaway early warning method for an energy storage power station according to any one of claims 1 to 3, characterized in that: The sensing data includes gas composition information measured by a gas composition detector, temperature information measured by a temperature sensor, and visible light information measured by a photoelectric smoke detector.

7. A battery thermal runaway warning device for an energy storage power station, characterized in that: include: A first training module is used to train a probabilistic neural network and a back propagation neural network using a training set until the probabilistic neural network converges, the training set includes a plurality of first training samples with hard labels, the training samples include sensor data collected by at least one sensor, and the hard label indicates whether a battery of the energy storage power station has thermal runaway or not; a soft label generation module, configured to use the converged probabilistic neural network to predict the second training sample without a label, and obtain a soft label of the second training sample, wherein the soft label represents a probability value of thermal runaway of the battery of the energy storage power station; A second training module, used for training the back propagation neural network using a plurality of second training samples with soft labels until the back propagation neural network converges; A risk prediction module is used to process the sensor data collected in real time from the energy storage power station by using the converged back propagation neural network to predict whether there is a thermal runaway risk in the battery of the energy storage power station; The runaway warning module is used to issue a thermal runaway warning when there is a risk of thermal runaway of the battery in the energy storage power station.

8. An electronic device, characterized in that: include: one or more processors; A storage device for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the battery thermal runaway early warning method for an energy storage power station as described in any one of claims 1-6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the battery thermal runaway warning method of the energy storage power station as described in any one of claims 1 to 6 is implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the computer program implements the battery thermal runaway early warning method for an energy storage power station as described in any one of claims 1 to 6.

Images