Electric power AI terminal dual-mode communication network and channel reliability evaluation method

Abstract

The invention discloses an electric power AI terminal dual-mode communication network and channel reliability evaluation method, and relates to the technical field of channel reliability data analysis, and the method comprises the following steps: collecting a plurality of historical signal strength values of the electric power AI terminal dual-mode communication network, dual-mode transmission mode data of a physical layer, and a bit error rate of a transmission channel; based on the plurality of historical signal strength values, the dual-mode transmission mode data of the physical layer and the bit error rate of the transmission channel, analyzing the predicted bit error rate of the transmission channel in different transmission modes; constructing a transmission channel reliability evaluation model; and evaluating the channel to be subjected to data transmission in real time through the constructed transmission channel reliability evaluation model, and performing real-time early warning on the transmission channel abnormity in the evaluation process. According to the method, the channel reliability evaluation model is constructed based on the historical signal strength, the bit error rate and the number of broken links, and the stability and reliability of a communication system in a complex environment are improved.

Description

Technical Field

[0001] This invention relates to the field of channel reliability data analysis technology, specifically to a dual-mode communication network for power AI terminals and a method for channel reliability assessment. Background Technology

[0002] With the deep integration of next-generation information and communication technologies and power systems, power AI terminals are widely deployed in scenarios such as power transmission and distribution, power consumption monitoring, and distribution network automation to achieve functions such as equipment status perception, environmental information collection, edge intelligent processing, and remote control. To ensure the high reliability and real-time communication of power AI terminals, communication networks typically adopt a dual-mode communication architecture, such as a combination of high-frequency carrier communication (HPLC) and high-frequency radio frequency communication (HRF), to enhance the network's anti-interference capability, transmission stability, and coverage. However, in the complex and ever-changing power field environment, the stability and reliability of the transmission channel face severe challenges due to factors such as electromagnetic interference, changes in line topology, and the mobility of communication nodes (e.g., distributed access or power outage reconnection). While HPLC has advantages such as strong penetration and the ability to achieve long-distance transmission using power lines, it is easily affected by load disturbances and transformer obstruction. HRF has the characteristics of flexible deployment and strong resistance to power interference, but signal attenuation is significant and reliability decreases in scenarios such as wall penetration and long distances. Therefore, in a dual-mode communication system, it is crucial to rationally select the optimal transmission mode and conduct real-time evaluation of the stability of the communication channel to improve overall network performance and ensure the security of power data transmission.

[0003] Traditional channel reliability assessment methods are mostly based on static threshold judgments or empirical rules, such as triggering alarms based on simple parameters like real-time RSSI (Received Signal Strength Indicator) and BER (Bit Error Rate), lacking the ability to perform deep learning and dynamic modeling of historical behavior patterns. Especially in dual-mode communication scenarios, the channel characteristics of the two modes differ significantly, making it impossible to uniformly judge communication quality using a single indicator. Furthermore, existing methods have low prediction accuracy when facing complex environmental disturbances (such as short-term interference, trunk line overload, and power grid oscillations), making it difficult to provide early warnings of potential link loss risks.

[0004] In recent years, with the widespread application of AI algorithms in communication systems, channel quality assessment methods based on machine learning or statistical modeling have gradually become a research hotspot. By modeling and fitting historical data such as signal strength, bit error rate, and number of communication link interruptions, the dynamic changes in the communication environment can be reflected more accurately, providing data support for real-time channel assessment and handover strategies.

[0005] Therefore, there is an urgent need for a new method that integrates the dual-mode communication architecture features of power AI terminals, can construct a channel reliability prediction model based on historical communication data, and achieve high-precision assessment and early warning of channel availability during actual communication. This method should comprehensively consider the bit error rate characteristics, signal strength variation trends, and link failure probabilities under different transmission modes, establish a mathematical model reflecting the correspondence between communication status and link failure risk, further guide network resource scheduling, improve communication security and efficiency, and provide stable support for business scenarios such as smart distribution networks and electricity monitoring. Summary of the Invention

[0006] The purpose of this invention is to provide a dual-mode communication network for power AI terminals and a method for evaluating channel reliability, so as to solve the problems raised in the prior art.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a dual-mode communication network for power AI terminals and a method for evaluating channel reliability, comprising the following steps: Step S100: Collect several historical signal strength values, physical layer dual-mode transmission mode data, and bit error rate of the transmission channel from the dual-mode communication network of the power AI terminal. Step S200: Based on several historical signal strength values, physical layer dual-mode transmission data, and the bit error rate of the transmission channel, analyze the predicted bit error rate of the transmission channel under different transmission modes. Step S300: Based on several signal strength values ​​of the dual-mode communication network of the power AI terminal and the bit error rate of the transmission channel under different transmission modes, construct a transmission channel reliability assessment model. Step S400: Evaluate the channel for real-time data transmission using the constructed transmission channel reliability assessment model, and provide real-time early warning for transmission channel anomalies during the assessment process.

[0008] Furthermore, in step S100, several historical signal strength values, physical layer dual-mode transmission mode data, and bit error rate of the transmission channel of the power AI terminal dual-mode communication network are collected, specifically: The set of signal strength values ​​of the source node during the data transmission process of the dual-mode communication network of the power AI terminal in history is: a={a1, a2, ..., au}; where u represents the total number of samples; Among them, the dual-mode transmission mode of the physical layer of the power AI terminal dual-mode communication network includes data reception and transmission using at least one of HPLC and HRF; the set of bit error rates of the transmission channel when data reception and transmission are performed using HPLC is: p={p1, p2, ..., pm}; where m represents the number of samples of bit error rates of the transmission channel when data reception and transmission are performed using HPLC in the history. The set of bit error rates of the transmission channel when data is received and transmitted using HRF is obtained as: q = {q1, q2, ..., qn}; where n represents the number of samples of bit error rates of the transmission channel when data is received and transmitted using HRF in the history. In the historical data transmission process of the dual-mode communication network of power AI terminals, the set of transmission channel disconnection times corresponding to the signal strength values ​​of different source nodes under different data reception and transmission methods is extracted as: f={f1, f2, ..., fu}.

[0009] It should be noted that this step is the data foundation of the entire method, collecting key parameters during the communication process: among which, signal strength (a): measures the strength of the signal transmitted by the source node and is an important factor affecting communication quality; dual-mode transmission (HPLC / HRF): used to distinguish the communication modes of the two physical transmission layers; bit error rate (BER, p / q): reflects the transmission accuracy of communication data and is an important indicator of channel quality; number of link breaks (f): represents abnormal interruption behavior during the communication process and is a direct reflection of the channel reliability assessment. Secondly, it should be noted that in this invention, the transmission channel disconnection takes into account the strength of the signal sent by the source node and the magnitude of the bit error rate. In practice, the factors that usually affect transmission channel disconnection are the strength of the signal sent by the source node and the bit error rate. Therefore, this invention is based on this practical problem. Other influencing factors, including terminal impact and transmission environment impact, are not considered in this solution because they are not obvious and have a small impact on transmission channel disconnection. That is, this invention is only applicable to situations where the terminal equipment and transmission environment are relatively stable. Furthermore, in step S200, based on several historical signal strength values, physical layer dual-mode transmission data, and the bit error rate of the transmission channel, the predicted bit error rate of the transmission channel under different transmission modes is analyzed, specifically as follows: The predicted bit error rate (B) of the transmission channel when data is received and transmitted using HPLC and the predicted bit error rate (D) of the transmission channel when data is received and transmitted using HRF are calculated using the average prediction method. The formula is: B = (p1 + p2 + ... + pm) / m; where p1, p2, ..., pm represent the bit error rate of the transmission channel when data is received and transmitted using HPLC in the 1st, 2nd, ..., mth times, respectively. D = (q1 + q2 + ... + qn) / n; where p1, p2, ..., pm represent the bit error rate of the transmission channel when data is received and sent using the HRF method for the 1st, 2nd, ..., nth times, respectively; The predicted bit error rate set for data reception and transmission using HPLC and HRF methods, calculated according to the above method, is: S = {S1, S2, ..., Su}; where set S includes the predicted bit error rate B of the transmission channel when data reception and transmission are performed using HPLC and the predicted bit error rate D of the transmission channel when data reception and transmission are performed using HRF.

[0010] Furthermore, in step S300, a transmission channel reliability assessment model is constructed based on several signal strength values ​​of the dual-mode communication network of the power AI terminal and the bit error rate of the transmission channel under different transmission modes, specifically as follows: The signal strength value a1 of the source node is calculated using the probability prediction formula P1=f1 / (f1+f2+...+fu), and the probability P1 of the transmission channel being disconnected when data is transmitted using the corresponding data receiving and transmitting methods is obtained; where fu represents the signal strength value au of the source node, and the number of times the transmission channel is disconnected when data is transmitted using the corresponding data receiving and transmitting methods. The probability set of transmission channel disconnection corresponding to the signal strength values ​​of different source nodes under different data reception and transmission methods is obtained by sequentially calculating the above calculation method: P={P1, P2, ..., Pu}; Based on the signal strength of the source node, the predicted bit error rate, and the probability of transmission channel disconnection, a triplet is formed: [(a1, S1, P1), (a2, S2, P2), ..., (au, Su, Pu)]. Using this triplet as a basis, real-time evaluation data for the transmission channel is collected. The collected real-time evaluation data is then fitted to construct a transmission channel reliability evaluation model, specifically characterized as follows: p' = α*x1 + β*x2 + γ; where α, β, and γ are regression coefficients; p' represents the fitting result term; and x1 and x2 represent the fitting parameters; The regression coefficients α, β, and γ are solved using the least squares method.

[0011] Furthermore, in step S400, the transmission channel reliability assessment model is used to assess the channel for real-time data transmission, and real-time early warning is given for transmission channel anomalies during the assessment process, specifically as follows: During the real-time data transmission process of the dual-mode communication network of the power AI terminal, the signal strength value a' of the source node is obtained; the dual-mode transmission mode of the physical layer of the current dual-mode communication network of the power AI terminal is obtained, and the predicted bit error rate S' is obtained. The signal strength value a' of the source node and the predicted bit error rate S' are then substituted into the constructed transmission channel reliability assessment model, specifically: Fitting parameters x1=a'; fitting parameters x2=S'; obtaining the fitting result term p'=α*a'+β*S'+γ; where the fitting result term p' represents the probability of transmission channel link failure, for a preset channel security threshold P'; when p'>P', the abnormal channel assessment result will be given a real-time warning.

[0012] Furthermore, the regression coefficients α, β, and γ are solved using the least squares method, specifically as follows: Based on the historical data transmission process of the dual-mode communication network of the power AI terminal, the set of source node signal strength values ​​is: a = {a1, a2, ..., au}. The set of predicted bit error rates for data reception and transmission using HPLC and HRF methods within the source node signal strength value set a is calculated as: S = {S1, S2, ..., Su}. The set of probabilities of transmission channel disconnection corresponding to different source node signal strength values ​​under different data reception and transmission methods is calculated as: P = {P1, P2, ..., Pu}. Regression coefficients are calculated using the least squares method. ; ; ; Where i represents the total number of samples, and j∈[1,u].

[0013] Furthermore, including: The reliability assessment model for transmission channels is fitted using computer programming, regression coefficients are solved, and the predicted bit error rate is calculated. It supports independent calculations by computer programs.

[0014] This invention is programmable and can be deployed in power AI terminals or edge computing devices to improve automation and intelligence levels. It facilitates later software system integration and embedded deployment, and is suitable for industrial-grade communication security early warning systems.

[0015] Compared with existing technologies, the beneficial effects of this invention are as follows: This invention provides a dual-mode communication network and channel reliability assessment method for power AI terminals. It can construct a channel reliability assessment model based on historical signal strength, bit error rate, and number of link failures, improving the stability and reliability of the communication system in complex environments. This method achieves accurate prediction of channel link failure probability through a fitted regression model, supports real-time early warning of communication anomalies, and effectively avoids data interruption caused by channel mutations. Simultaneously, the method supports adaptive evaluation and optimization switching between HPLC and HRF dual-mode communication modes, providing a decision-making basis for intelligent scheduling of communication methods. The model uses highly interpretable least-squares regression modeling, is computationally simple, and is easy to deploy on power edge terminals, possessing good practicality and promotional value. Furthermore, this method has strong compatibility and scalability, is applicable to various power communication architectures, and can effectively support the intelligent development of power IoT communication systems. Attached Figure Description

[0016] Figure 1 This is a schematic diagram illustrating the steps of a dual-mode communication network and channel reliability assessment method for a power AI terminal according to the present invention. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] Example: Figure 1 As shown, the present invention provides a technical solution, a dual-mode communication network for power AI terminals and a method for evaluating channel reliability, including the following steps: Step S100, collecting several historical signal strength values, physical layer dual-mode transmission mode data and bit error rate of the transmission channel of the dual-mode communication network for power AI terminals; The set of signal strength values ​​of the source node during the data transmission process of the dual-mode communication network of the power AI terminal in history is: a={a1, a2, ..., au}; where u represents the total number of samples; Among them, the dual-mode transmission mode of the physical layer of the power AI terminal dual-mode communication network includes data reception and transmission using at least one of HPLC and HRF; the set of bit error rates of the transmission channel when data reception and transmission are performed using HPLC is: p={p1, p2, ..., pm}; where m represents the number of samples of bit error rates of the transmission channel when data reception and transmission are performed using HPLC in the history. The set of bit error rates of the transmission channel when data is received and transmitted using HRF is obtained as: q = {q1, q2, ..., qn}; where n represents the number of samples of bit error rates of the transmission channel when data is received and transmitted using HRF in the history. In the historical data transmission process of the dual-mode communication network of power AI terminals, the set of transmission channel disconnection times corresponding to the signal strength values ​​of different source nodes under different data reception and transmission methods is extracted as: f={f1, f2, ..., fu}.

[0019] Step S200: Based on several historical signal strength values, physical layer dual-mode transmission data, and the bit error rate of the transmission channel, analyze the predicted bit error rate of the transmission channel under different transmission modes. The predicted bit error rate (B) of the transmission channel when data is received and transmitted using HPLC and the predicted bit error rate (D) of the transmission channel when data is received and transmitted using HRF are calculated using the average prediction method. The formula is: B = (p1 + p2 + ... + pm) / m; where p1, p2, ..., pm represent the bit error rate of the transmission channel when data is received and transmitted using HPLC in the 1st, 2nd, ..., mth times, respectively. D = (q1 + q2 + ... + qn) / n; where p1, p2, ..., pm represent the bit error rate of the transmission channel when data is received and sent using the HRF method for the 1st, 2nd, ..., nth times, respectively; The predicted bit error rate set for data reception and transmission using HPLC and HRF methods, calculated according to the above method, is: S = {S1, S2, ..., Su}; where set S includes the predicted bit error rate B of the transmission channel when data reception and transmission are performed using HPLC and the predicted bit error rate D of the transmission channel when data reception and transmission are performed using HRF.

[0020] Step S300: Based on several signal strength values ​​of the dual-mode communication network of the power AI terminal and the bit error rate of the transmission channel under different transmission modes, construct a transmission channel reliability assessment model. The signal strength value a1 of the source node is calculated using the probability prediction formula P1=f1 / (f1+f2+...+fu), and the probability P1 of the transmission channel being disconnected when data is transmitted using the corresponding data receiving and transmitting methods is obtained; where fu represents the signal strength value au of the source node, and the number of times the transmission channel is disconnected when data is transmitted using the corresponding data receiving and transmitting methods. The probability set of transmission channel disconnection corresponding to the signal strength values ​​of different source nodes under different data reception and transmission methods is obtained by sequentially calculating the above calculation method: P={P1, P2, ..., Pu}; Based on the signal strength of the source node, the predicted bit error rate, and the probability of transmission channel disconnection, a triplet is formed: [(a1, S1, P1), (a2, S2, P2), ..., (au, Su, Pu)]. Using this triplet as a basis, real-time evaluation data for the transmission channel is collected. The collected real-time evaluation data is then fitted to construct a transmission channel reliability evaluation model, specifically characterized as follows: p' = α*x1 + β*x2 + γ; where α, β, and γ are regression coefficients; p' represents the fitting result term; and x1 and x2 represent the fitting parameters; The regression coefficients α, β, and γ are solved using the least squares method.

[0021] Furthermore, the regression coefficients α, β, and γ are solved using the least squares method, specifically as follows: Based on the historical data transmission process of the dual-mode communication network of the power AI terminal, the set of source node signal strength values ​​is: a = {a1, a2, ..., au}. The set of predicted bit error rates for data reception and transmission using HPLC and HRF methods within the source node signal strength value set a is calculated as: S = {S1, S2, ..., Su}. The set of probabilities of transmission channel disconnection corresponding to different source node signal strength values ​​under different data reception and transmission methods is calculated as: P = {P1, P2, ..., Pu}. Regression coefficients are calculated using the least squares method. ; ; ; Where i represents the total number of samples, and j∈[1,u].

[0022] Among them, the least squares method is an existing technique that is widely used to solve regression coefficients; It should be noted that this step is the data foundation of the entire method, collecting key parameters during the communication process: among which, signal strength (a): measures the strength of the signal transmitted by the source node and is an important factor affecting communication quality; dual-mode transmission (HPLC / HRF): used to distinguish the communication modes of the two physical transmission layers; bit error rate (BER, p / q): reflects the transmission accuracy of communication data and is an important indicator of channel quality; number of link breaks (f): represents abnormal interruption behavior during the communication process and is a direct reflection of the channel reliability assessment. Secondly, it should be noted that in this invention, the transmission channel disconnection takes into account the strength of the signal sent by the source node and the magnitude of the bit error rate. In practice, the factors that usually affect transmission channel disconnection are the strength of the signal sent by the source node and the bit error rate. Therefore, this invention is based on this practical problem. Other influencing factors, including terminal impact and transmission environment impact, are not considered in this solution because they are not obvious and have a small impact on transmission channel disconnection. That is, this invention is only applicable to situations where the terminal equipment and transmission environment are relatively stable. Step S400: Evaluate the channel for real-time data transmission using the constructed transmission channel reliability assessment model, and provide real-time early warning for transmission channel anomalies during the assessment process.

[0023] During the real-time data transmission process of the dual-mode communication network of the power AI terminal, the signal strength value a' of the source node is obtained; the dual-mode transmission mode of the physical layer of the current dual-mode communication network of the power AI terminal is obtained, and the predicted bit error rate S' is obtained. The signal strength value a' of the source node and the predicted bit error rate S' are then substituted into the constructed transmission channel reliability assessment model, specifically: Fitting parameters x1=a'; fitting parameters x2=S'; obtaining the fitting result term p'=α*a'+β*S'+γ; where the fitting result term p' represents the probability of transmission channel link failure, for a preset channel security threshold P'; when p'>P', the abnormal channel assessment result will be given a real-time warning.

[0024] As a preferred embodiment Source node signal strength set (unit: dBm): a = {-72, -68, -80, -65, -75} (sample size u = 5); HPLC mode bit error rate set: p={3.2*10 -5 2.8*10 -5 4.1*10 -5 (Sample size m=3); HRF mode bit error rate set: q={7.5*10 -4 6.3*10 -4 8.1*10 -4 9.0*10 -4 (Sample size n=4); Set of chain break counts (corresponding to signal strength a): f={2,1,5,0,3} (the number of times the link was broken in 5 transmissions); HPLC prediction error rate: B = 3.37 * 10 -5; HRF prediction bit error rate: D = 7.73 * 10 -4 ; Generate a set of bit error rates (matched by transmission mode): S={3.37*10 -5 7.73*10 -4 3.37*10 -5 7.73*10 -4 3.37*10 -5}; Calculate the probability of link failure (using 100 total transmissions as a baseline): P = {0.02, 0.01, 0.05, 0, 0.03}; Composing a triplet dataset: [(-72,3.37*10 -5 ,0.02), (-68, 7.73*10) -4 ,0.01), (-80, 3.37*10) -5 ,0.05), (-65, 7.73*10) -4 ,0), (-75, 3.37*10) -5 [,0.03)]; The regression coefficients are obtained using the least squares method: α=0.0012, β=28.5, γ=0.15; The channel reliability assessment model is obtained as follows: p' = 0.0012 * a + 28.5 * S + 0.15; In real-time data, Signal strength a' = -78 dBm; Using HPLC mode → Predicted bit error rate S' = 3.37 * 10 -5 ; Substitute into the model for calculation: p'=0.0012*(-78)+28.5*(3.37*10 -5 ) + 0.15 ≈ 0.057; Set the safety threshold P'=0.05, Comparison results: 0.057 > 0.05 → Triggers channel anomaly warning; Furthermore, including: The reliability assessment model for transmission channels is fitted using computer programming, regression coefficients are solved, and the predicted bit error rate is calculated. It supports independent calculations by computer programs.

[0025] This invention is programmable and can be deployed in power AI terminals or edge computing devices to improve automation and intelligence levels. It facilitates later software system integration and embedded deployment, and is suitable for industrial-grade communication security early warning systems.

[0026] Furthermore, this invention can also realize a method for predicting the reliability of dual-mode communication channels in power AI terminals through machine learning, and greatly improve training cost and training time through feature-based training. Secondly, this method uses a data-driven approach to predict the probability of channel disconnection and achieve real-time early warning. By extracting features, training models, and making online predictions from historical dual-mode communication data, the stability and security of the communication channel are improved.

[0027] Historical data from the dual-mode communication network of the power AI terminal was collected, including: The set of signal strength values ​​of the source node: a = {a1, a2, ..., au}; where u represents the total number of samples; Dual-mode physical layer transport (HPLC and HRF); Bit error rate in different modes: The set of bit error rates in HPLC mode: p = {p1, p2, ..., pm}; where m represents the number of samples of bit error rates of the transmission channel when data is received and transmitted using HPLC in the historical data acquisition process; The set of bit error rates in HRF mode is: q = {q1, q2, ..., qn}; where n represents the number of samples of bit error rates of the transmission channel when data is received and sent using HRF mode in the history. Set of communication link disconnection counts: f = {f1, f2, ..., fu}; By cleaning and normalizing the raw data, a training sample set is formed: the data in the training sample set is the historical data of the dual-mode communication network of the power AI terminal collected above, i.e., triples. The normalization method is the Max-Min normalization method. For different transmission modes, a combination of statistical and machine learning methods is used to calculate the predicted bit error rate, which is represented by the formula: B = (p1 + p2 + ... + pm) / m; where p1, p2, ..., pm represent the bit error rate of the transmission channel when data is received and transmitted using the HPLC method for the 1st, 2nd, ..., mth times, respectively. D = (q1 + q2 + ... + qn) / n; where p1, p2, ..., pm represent the bit error rate of the transmission channel when data is received and sent using the HRF method for the 1st, 2nd, ..., nth times, respectively; The average predicted value of the model is added to the sample set as a feature term to form a predicted feature set S={S1, S2, ..., Su}, and classification is completed according to the transport mode by HPLC and HRF.

[0028] A transmission channel reliability prediction model is constructed, with signal strength and predicted bit error rate as input features and link failure probability as output variable, forming supervised learning samples: lightweight machine learning models (LightGBM, XGBoost, SVR) are selected for fitting, p'=α*a'+β*S'+γ. The training parameters are solved using the least squares method.

[0029] Loss control is achieved by minimizing the mean square error of u training samples.

[0030] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

Claims

1. A dual-mode communication network for power AI terminals and a method for evaluating channel reliability, characterized in that: Includes the following steps: Step S100: Collect several historical signal strength values, physical layer dual-mode transmission mode data, and bit error rate of the transmission channel from the dual-mode communication network of the power AI terminal. Step S200: Based on several historical signal strength values, physical layer dual-mode transmission data, and the bit error rate of the transmission channel, analyze the predicted bit error rate of the transmission channel under different transmission modes. Step S300: Based on several signal strength values ​​of the dual-mode communication network of the power AI terminal and the bit error rate of the transmission channel under different transmission modes, construct a transmission channel reliability assessment model. Step S400: Evaluate the channel for real-time data transmission using the constructed transmission channel reliability assessment model, and provide real-time early warning for transmission channel anomalies during the assessment process.

2. The dual-mode communication network and channel reliability assessment method for power AI terminals according to claim 1, characterized in that: In step S100, several historical signal strength values, physical layer dual-mode transmission mode data, and bit error rate of the transmission channel of the power AI terminal dual-mode communication network are collected, specifically: The set of signal strength values ​​of the source node during the data transmission process in the dual-mode communication network of the power AI terminal in history is: a = {a1, a2, ..., au}; where u represents the total number of samples; Among them, the dual-mode transmission mode of the physical layer of the power AI terminal dual-mode communication network includes using at least one of HPLC and HRF for data reception and transmission.

3. The dual-mode communication network and channel reliability assessment method for power AI terminals according to claim 2, characterized in that: include: The set of bit error rates of the transmission channel when data is received and transmitted using HPLC is obtained as: p = {p1, p2, ..., pm}; where m represents the number of samples of bit error rates of the transmission channel when data is received and transmitted using HPLC in the history. The set of bit error rates of the transmission channel when data is received and transmitted using HRF is obtained as: q = {q1, q2, ..., qn}; where n represents the number of samples of bit error rates of the transmission channel when data is received and transmitted using HRF in the history. In the historical data transmission process of the dual-mode communication network of power AI terminals, the set of transmission channel disconnection times corresponding to the signal strength values ​​of different source nodes under different data reception and transmission methods is extracted as: f={f1, f2, ..., fu}.

4. The dual-mode communication network and channel reliability assessment method for power AI terminals according to claim 3, characterized in that: In step S200, based on several historical signal strength values, physical layer dual-mode transmission data, and the bit error rate of the transmission channel, the predicted bit error rate of the transmission channel under different transmission modes is analyzed, specifically as follows: The predicted bit error rate (B) of the transmission channel when using HPLC for data reception and transmission and the predicted bit error rate (D) of the transmission channel when using HRF for data reception and transmission are calculated using the average prediction method.

5. The dual-mode communication network and channel reliability assessment method for power AI terminals according to claim 4, characterized in that: Specifically: The formula for the predicted bit error rate B of the transmission channel when using HPLC for data reception and transmission is: B = (p1 + p2 + ... + pm) / m; where p1, p2, ..., pm represent the bit error rate of the transmission channel when using HPLC for data reception and transmission for the 1st, 2nd, ..., mth times, respectively. The predicted bit error rate of the transmission channel when using HRF for data reception and transmission is D = (q1 + q2 + ... + qn) / n; where p1, p2, ..., pm represent the bit error rate of the transmission channel when using HRF for data reception and transmission for the 1st, 2nd, ..., nth times, respectively. The predicted bit error rate set for data reception and transmission using HPLC and HRF methods, calculated according to the above method, is: S = {S1, S2, ..., Su}; where set S includes the predicted bit error rate B of the transmission channel when data reception and transmission are performed using HPLC and the predicted bit error rate D of the transmission channel when data reception and transmission are performed using HRF.

6. The dual-mode communication network and channel reliability assessment method for power AI terminals according to claim 5, characterized in that: In step S300, a transmission channel reliability assessment model is constructed based on several signal strength values ​​of the dual-mode communication network of the power AI terminal and the bit error rate of the transmission channel under different transmission modes. Specifically: The signal strength value a1 of the source node is calculated using the probability prediction formula P1=f1 / (f1+f2+...+fu), and the probability P1 of the transmission channel being disconnected when data is transmitted using the corresponding data receiving and transmitting methods is obtained; where fu represents the signal strength value au of the source node, and the number of times the transmission channel is disconnected when data is transmitted using the corresponding data receiving and transmitting methods. The probability set of transmission channel disconnection corresponding to the signal strength values ​​of different source nodes under different data reception and transmission methods is obtained by sequentially calculating the above calculation method: P={P1, P2, ..., Pu}; Based on the signal strength of the source node, the predicted bit error rate, and the probability of transmission channel disconnection, a triplet is formed: [(a1, S1, P1), (a2, S2, P2), ..., (au, Su, Pu)]. Using this triplet as a basis, real-time evaluation data for the transmission channel is collected. The collected real-time evaluation data is then fitted to construct a transmission channel reliability evaluation model, specifically characterized as follows: p' = α*x1 + β*x2 + γ; where α, β, and γ are regression coefficients; p' represents the fitting result term; and x1 and x2 represent the fitting parameters; The regression coefficients α, β, and γ are solved using the least squares method.

7. The dual-mode communication network and channel reliability assessment method for a power AI terminal according to claim 6, characterized in that: In step S400, the transmission channel reliability assessment model is used to assess the channel for real-time data transmission, and real-time early warning is given for transmission channel anomalies during the assessment process. Specifically: During the real-time data transmission process of the dual-mode communication network of the power AI terminal, the signal strength value a' of the source node is obtained; the dual-mode transmission mode of the physical layer of the current dual-mode communication network of the power AI terminal is obtained, and the predicted bit error rate S' is obtained. The signal strength value a' of the source node and the predicted bit error rate S' are then substituted into the constructed transmission channel reliability assessment model, specifically: Fitting parameters x1=a'; fitting parameters x2=S'; obtaining the fitting result term p'=α*a'+β*S'+γ; where the fitting result term p' represents the probability of transmission channel link failure, for a preset channel security threshold P'; when p'>P', the abnormal channel assessment result will be given a real-time warning.

8. The dual-mode communication network and channel reliability assessment method for power AI terminals according to claim 6, characterized in that: The regression coefficients α, β, and γ are solved using the least squares method, specifically as follows: Based on the historical data transmission process of the dual-mode communication network of the power AI terminal, the set of source node signal strength values ​​is: a = {a1, a2, ..., au}. The set of predicted bit error rates for data reception and transmission using HPLC and HRF methods within the source node signal strength value set a is calculated as: S = {S1, S2, ..., Su}. The set of probabilities of transmission channel disconnection corresponding to different source node signal strength values ​​under different data reception and transmission methods is calculated as: P = {P1, P2, ..., Pu}. Regression coefficients are calculated using the least squares method. ； ； ； Where i represents the total number of samples, and j∈[1,u].

9. The dual-mode communication network and channel reliability assessment method for a power AI terminal according to claim 8, characterized in that: include: The reliability assessment model for transmission channels is fitted using computer programming, regression coefficients are solved, and the predicted bit error rate is calculated. It supports independent calculations by computer programs.

Images