Service capability query method based on Internet of Things and website management center system

Abstract

The invention discloses a service capability query method based on the Internet of Things and a website management center system, an Internet of Things sensor array and a data acquisition terminal are arranged outside a financial service facility, real-time environment and equipment state data are collected based on the Internet of Things sensor array and the data acquisition terminal, and a service capability query model is established; based on the established service capability query model, putting real-time data into a risk assessment model, and determining an optimal service response time parameter; and optimizing the service capability query model based on an optimal service response time parameter and an optimal query strategy, adjusting the data acquisition terminal based on the optimized model, and establishing a preset risk monitoring and early warning system. The method breaks through the limitation of a traditional fixed threshold value, intelligently adjusts a service response time critical value in real time, remarkably shortens service delay, reduces the occurrence rate of extreme delay events, intelligently shunts high-load requests through distributed nodes, guarantees the operation stability of a system, and achieves zero-abnormality restart of key equipment.

Description

Technical Field

[0001] The present invention relates to the field of Internet of Things financial security, and in particular to a service capability query method and system based on the Internet of Things. Background Art

[0002] Current branch management center systems generally suffer from sluggish response times and a lack of risk awareness. Traditional solutions rely on static threshold mechanisms, such as fixed response timeout thresholds and periodic equipment inspections, which are difficult to cope with the highly dynamic environment changes in financial venues. For example:

[0003] Delayed response to environmental interference: Sudden peaks in passenger flow or sudden changes in temperature and humidity (such as electromagnetic interference caused by heavy rain) can easily cause service terminal CPU overload or network packet loss. The existing system lacks the ability to quantify environmental factors in real time.

[0004] The hardware status is disconnected from the business load: Hardware failures such as excessive ATM vibration or a cash box running out of money only trigger independent alarms and are not linked to real-time business queues for decision-making, resulting in extended fault recovery time during peak periods; a slight single defect: When the network is congested, it only relies on the basic caching mechanism and cannot coordinate the execution of edge node diversion and dynamic encryption, resulting in a high standard deviation of response time fluctuations.

[0005] The industry has attempted to introduce machine learning to optimize response models, but there are two limitations: missing feature dimensions: the time series correlation between device health and business peaks is not integrated; and insufficient closed-loop optimization: the policy execution results are not fed back into the risk assessment model iteration, resulting in a high model drift rate.

[0006] Therefore, it is a problem worth studying to provide a service capability query method and branch management center system that can well serve financial institutions by taking advantage of the existing Internet of Things. Summary of the Invention

[0007] In order to solve the above-mentioned deficiencies in the prior art, the present invention provides a service capability query method and a network management center system based on the Internet of Things.

[0008] The purpose of the present invention is to achieve:

[0009] The service capability query method based on the Internet of Things includes:

[0010] An IoT sensor array and a data collection terminal are set up outside the financial service facility, and real-time environment and equipment status data are collected based on the IoT sensor array and the data collection terminal to establish a service capability query model;

[0011] Based on the established service capability query model, real-time data is put into the risk assessment model to determine the optimal service response time parameters;

[0012] After selecting the optimal service response time parameters, put the real-time data and the optimal parameters into the query optimization model to determine the optimal query strategy;

[0013] The service capability query model is optimized based on the preferred service response time parameter and the preferred query strategy, the data collection terminal is adjusted based on the optimized model, and a preset risk monitoring and early warning system is established.

[0014] As a preferred solution of the service capability query method based on the Internet of Things of the present invention, the service capability query model includes:

[0015] Comprehensive service quality index algorithm:

[0016]

[0017] Among them, α, β, γ are weight factors, ranging from [0, 1], and satisfying α + β + γ = 1. The initial value of the weight is preset by the system. S is the comprehensive service quality index, and T is the service response time. max Refers to the maximum threshold of service response time, N is the network load factor, C is the environmental interference factor, C base It is the baseline value of the environmental interference factor and the steady-state reference value set based on historical data.

[0018] As a preferred solution of the service capability query method based on the Internet of Things of the present invention, the Internet of Things sensor array includes:

[0019] Vibration sensors and cash box status sensors deployed in ATM equipment;

[0020] CPU temperature sensor and network delay detection module installed on the service terminal;

[0021] Temperature and humidity sensors and crowd density cameras installed in business premises.

[0022] As a preferred solution of the service capability query method based on the Internet of Things of the present invention, the risk assessment model is a dynamic decision tree model, and its input layer includes:

[0023] Equipment failure warning level based on equipment health classification;

[0024] Real-time business request peaks from service queue monitoring;

[0025] Environmental safety score composed of environmental interference and security system;

[0026] When determining the optimal service response time parameter, the service response time prediction formula is used:

[0027] T=a×D+b×N+c×C+d

[0028] Where D represents the device health indicator, a, b, and c are weight coefficients obtained through historical training, and d is the baseline response time constant.

[0029] As a preferred solution of the service capability query method based on the Internet of Things of the present invention, the determination of the preferred query strategy specifically includes:

[0030] When T>threshold T_max, distributed edge nodes are enabled to offload query requests;

[0031] When the network load factor N>0.8, the query result cache priority mechanism is triggered;

[0032] Based on the mutation amplitude of the environmental interference factor C, the data encryption level is dynamically adjusted.

[0033] In view of the above problems and / or the problems existing in the existing service capability query methods based on the Internet of Things, the present invention is proposed.

[0034] Therefore, the problem to be solved by the present invention is:

[0035] Establish an effective financial service system based on the Internet of Things.

[0036] To solve the above technical problems, the present invention provides the following technical solutions: a network point management center system based on the Internet of Things, comprising:

[0037] Environmental perception module, risk decision module, query execution module, terminal interaction module and strategy feedback module;

[0038] The environmental perception module includes a device monitoring unit and a data acquisition unit. The device monitoring unit is used to obtain hardware status data of ATM devices and service terminals in real time. The data acquisition unit collects environmental and network load data through temperature and humidity sensors and network probes.

[0039] The risk decision module dynamically optimizes the service response time parameters through the risk assessment model and transmits the parameter instructions to the query execution module;

[0040] The terminal interaction module allows the user to submit a service capability query request through a mobile terminal or a voice interface, and transmits the request to the query execution module;

[0041] The query execution module includes a strategy allocation unit and an optimization unit. The strategy allocation unit selects a query path based on the output instruction of the risk decision module, and the optimization unit dynamically adjusts resource allocation according to the prediction formula.

[0042] The strategy feedback module records the execution status of all query requests, summarizes the response delay data and environmental variables, and feeds them back to the risk decision module for model iterative optimization.

[0043] As a preferred solution of the network management center system based on the Internet of Things of the present invention, wherein: the equipment monitoring unit integrates a vibration sensor, a cash box status sensor and a CPU temperature sensor;

[0044] The data acquisition unit deploys an electromagnetic interference detector and a three-dimensional anemometer;

[0045] The terminal interaction module supports natural language parsing and converts voice requests into structured query instructions.

[0046] As a preferred solution of the network point management center system based on the Internet of Things of the present invention, the operation of the optimization unit includes:

[0047] When the predicted response time T> threshold T max When , the distributed edge nodes are activated to divert query requests;

[0048] Dynamically switch the data encryption protocol level based on the numerical range of the environmental interference factor;

[0049] A compensation strategy for the network load factor is generated through the dynamic decision tree model.

[0050] The present invention provides the following technical solution: an electronic device comprising:

[0051] one or more processors;

[0052] a storage device having one or more programs stored thereon;

[0053] When the one or more programs are executed by the one or more processors, the one or more processors implement a service capability query method based on the Internet of Things.

[0054] The present invention provides the following technical solution: an electronic device comprising:

[0055] A computer-readable storage medium stores executable instructions, which, when executed by a processor, enable the processor to implement a service capability query method based on the Internet of Things.

[0056] Positive and beneficial effects: The present invention proposes a service capability query method and an outlet management center system based on the Internet of Things, which realizes real-time and precise control of service capabilities through multi-dimensional environmental perception and intelligent decision-making models. The system breaks through the traditional fixed threshold limit and intelligently adjusts the service response time critical value in real time, significantly shortening business delays and reducing the incidence of extreme delay events. At the same time, it automatically activates the data security reinforcement and service diversion mechanism in the event of abnormal environmental mutations such as strong electromagnetic interference or peak traffic flow, compressing the service interruption time to minutes, greatly improving the resilience of the system. For the first time, the integration of factors such as equipment operating status, business load pressure and environmental safety risks for full-factor collaborative monitoring significantly improves the accuracy of equipment fault warning and business request processing efficiency, and uses closed-loop feedback to achieve minute-level iteration of the decision model to completely solve the problem of model aging. During the environmental stability period, intelligent throttling technology is used to reduce the intensity of data collection, effectively saving network transmission and storage resources, and intelligently diverting high-load requests through distributed nodes to ensure the smooth operation of the system and achieve zero-abnormal restart of key equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 This is an operational flow chart of the service capability query method based on the Internet of Things in Example 1.

[0058] Figure 2 This is an operational flow chart of the network management center system based on the Internet of Things in Example 2. DETAILED DESCRIPTION

[0059] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0060] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0061] Secondly, the term "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in various places throughout this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive of other embodiments.

[0062] Example 1

[0063] Reference Figure 1 , which is the first embodiment of the present invention, provides a service capability query method based on the Internet of Things, which includes:

[0064] An IoT sensor array and a data collection terminal are set up outside the financial service facility, and real-time environment and equipment status data are collected based on the IoT sensor array and the data collection terminal to establish a service capability query model;

[0065] Based on the established service capability query model, real-time data is put into the risk assessment model to determine the optimal service response time parameters;

[0066] Comprehensive service quality index algorithm:

[0067]

[0068] Among them, α, β, γ are weight factors, ranging from [0, 1], and satisfying α + β + γ = 1. The initial value of the weight is preset by the system. S is the comprehensive service quality index, and T is the service response time. max Refers to the maximum threshold of service response time, N is the network load factor, C is the environmental interference factor, C base It is the baseline value of the environmental interference factor and the steady-state reference value set based on historical data.

[0069] After selecting the optimal service response time parameters, put the real-time data and the optimal parameters into the query optimization model to determine the optimal query strategy;

[0070] The IoT sensor array includes:

[0071] Vibration sensors and cash box status sensors deployed in ATM equipment;

[0072] CPU temperature sensor and network delay detection module installed on the service terminal;

[0073] Temperature and humidity sensors and crowd density cameras installed in business premises.

[0074] The service capability query model is optimized based on the preferred service response time parameter and the preferred query strategy, the data collection terminal is adjusted based on the optimized model, and a preset risk monitoring and early warning system is established.

[0075] The risk assessment model is a dynamic decision tree model, and its input layer includes:

[0076] Equipment failure warning level based on equipment health classification;

[0077] Real-time business request peaks from service queue monitoring;

[0078] Environmental safety score composed of environmental interference and security system;

[0079] When determining the optimal service response time parameter, the service response time prediction formula is used:

[0080] T=a×D+b×N+c×C+d

[0081] Where D represents the device health indicator, a, b, and c are weight coefficients obtained through historical training, and d is the baseline response time constant.

[0082] The determination of the optimal query strategy specifically includes:

[0083] When T>threshold T_max, distributed edge nodes are enabled to offload query requests;

[0084] When the network load factor N>0.8, the query result cache priority mechanism is triggered;

[0085] Based on the mutation amplitude of the environmental interference factor C, the data encryption level is dynamically adjusted.

[0086] Implementation example of the branch management center system based on the Internet of Things:

[0087] Implementation environment configuration:

[0088] A bank branch deployed an IoT system, including:

[0089] Sensor Array:

[0090] The ATM is equipped with a vibration sensor (sensitivity ±0.1G) and a cash box status sensor (capacity detection accuracy ±5 sheets).

[0091] The service terminal integrates a CPU temperature sensor (range 0-100°C) and a network delay detection module (sampling rate 1 time / second).

[0092] Temperature and humidity sensors (temperature error ±0.5°C, humidity error ±3%) and AI crowd density cameras (recognition accuracy >95%) are deployed on the ceiling of the business hall.

[0093] Data collection terminal: edge computing gateway (industrial-grade ARM processor, supporting 5G backhaul).

[0094] Step 1: Establish a service capability query model:

[0095] Data collection example (data at 10:00 on June 5, 2025):

[0096] Device status: ATM003 vibration value 0.3G (normal threshold <0.5G), banknote balance 23% → device health D = 0.85.

[0097] Network status: Bandwidth utilization 78%, packet loss rate 2% → Network load factor N = 0.78 × 0.7 + (2 / 100) × 0.3 = 0.552.

[0098] Environmental data: temperature 28°C, humidity 65%, electromagnetic interference 1.2 V / m → environmental interference factor C = 0.4×|28-25| / 10+0.3×|65-60| / 20+0.3×1.2=0.42.

[0099] Model initialization:

[0100] The initial weights of the comprehensive service quality index S are: α = 0.5 (response time weight), β = 0.3 (network weight), and γ = 0.2 (environment weight).

[0101] Benchmark parameter: T max =1500ms (maximum tolerance response time), C base =0.35 (historical environmental interference benchmark).

[0102] Step 2: Risk assessment and parameter optimization:

[0103] Dynamic decision tree input:

[0104] Equipment warning level: D = 0.85 → Level 1 (0.8-1.0 is safe).

[0105] Business request peak: 52 concurrent queries → TS = log1(52+1) = 1.73.

[0106] Environmental safety score: EI = 0.42 > C base =0.35→Safety score=70 (percentage point).

[0107] Response time prediction calculation:

[0108] Input: D = 0.85, N = 0.552, C = 0.42 (simplified value of C).

[0109] Weight coefficients: a=1.2, b=0.8, c=0.5 (historical training values), d=200ms.

[0110] Output: T = 1.2 × 0.85 + 0.8 × 0.552 + 0.5 × 0.42 + 200 = 201.8 ms.

[0111] Step 3: Query policy execution:

[0112] Strategy trigger logic:

[0113] Because T = 201.8ms <T max = 1500ms → Edge offloading is not enabled.

[0114] Network load N = 0.552 < 0.8 → Disable the cache priority mechanism.

[0115] Environmental interference change rate ΔC = |0.42-0.35| / 10 min = 0.007 / min < threshold 0.01 → AES-128 encryption is maintained.

[0116] Real-time quality of service calculation:

[0117] S=0.5×(1-201.8 / 1500)+0.3×(1-0.552)+0.2×(1-0.42 / 0.35)=0.82>the excellent threshold of 0.7, generating the "adequate service capacity" status code.

[0118] Step 4: Model optimization and risk warning:

[0119] Exception handling example (sudden peak in traffic at 10:30):

[0120] The crowd density camera detects >20 people in the waiting area → triggers model recalibration.

[0121] Automatically adjust weights: α = 0.4, β = 0.4, γ = 0.2 (increase the weight of network factors).

[0122] New warning: When S<0.6, a "service degradation alert" is pushed to the administrator.

[0123] Historical data iteration:

[0124] A total of 1,842 queries were executed that day, and model parameters were dynamically updated:

[0125] Retraining the random forest regression model with weights a, b, c (dataset expanded by 3.2%).

[0126] The benchmark value C_base was revised from 0.35 to 0.38 (reflecting the change in the summer environmental baseline).

[0127] Key Data Flow Verification Table

[0128]

[0129] Example 2

[0130] Reference Figure 2 , which is a second embodiment of the present invention, the network management center system based on the Internet of Things includes:

[0131] Environmental perception module 100, risk decision module 200, query execution module 300, terminal interaction module 400 and strategy feedback module 500;

[0132] The environment perception module 100 includes a device monitoring unit 101 and a data acquisition unit 102. The device monitoring unit 101 is used to obtain hardware status data of ATM devices and service terminals in real time. The data acquisition unit 102 collects environmental and network load data through temperature and humidity sensors and network probes.

[0133] The equipment monitoring unit 101 integrates a vibration sensor, a cash box status sensor, and a CPU temperature sensor;

[0134] The data acquisition unit 102 deploys an electromagnetic interference detector and a three-dimensional anemometer;

[0135] The terminal interaction module 300 supports natural language analysis and converts voice requests into structured query instructions.

[0136] The risk decision module 200 dynamically optimizes the service response time parameters through the risk assessment model and transmits the parameter instructions to the query execution module 300;

[0137] The terminal interaction module 400 allows the user to submit a service capability query request through a mobile terminal or a voice interface, and transmits the request to the query execution module 300;

[0138] The query execution module 300 includes a strategy allocation unit 301 and an optimization unit 302. The strategy allocation unit 301 selects a query path based on the output instruction of the risk decision module 200, and the optimization unit 302 dynamically adjusts resource allocation according to the prediction formula.

[0139] The operations of the optimization unit 302 include:

[0140] When the predicted response time T> threshold T max When , the distributed edge nodes are activated to divert query requests;

[0141] Dynamically switch the data encryption protocol level based on the numerical range of the environmental interference factor;

[0142] A compensation strategy for the network load factor is generated through the dynamic decision tree model.

[0143] The policy feedback module 500 records the execution status of all query requests, summarizes the response delay data and environmental variables, and feeds them back to the risk decision module 200 for model iterative optimization.

[0144] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

[0145] Example 3

[0146] A third embodiment of the present invention is a network management center system based on the Internet of Things, including:

[0147] Test preparation and specific implementation process:

[0148] A 30-day comparative test was conducted at a Bank of China branch in Shanghai, selecting Business Hall 1 (experimental group) and Business Hall 2 (control group), both of which have similar physical environments. The experimental group fully deployed the system:

[0149] Environmental perception module: Piezoelectric vibration sensors (range ±5g, accuracy 0.02g) and weighing cash box sensors (error ±0.2%) are embedded in 5 ATMs; 12 intelligent teller machines are equipped with infrared CPU temperature probes (±0.5°C) and dual-band network probes (delay detection resolution 1ms); 8 temperature and humidity composite sensors (temperature range -40°C to 125°C, humidity range 0-100% RH) and 6 AI crowd counting cameras (detection accuracy 98%, coverage radius 8 meters) are installed in the business area.

[0150] Data collection terminal: Using an industrial-grade Raspberry Pi 4B cluster, sensor data is synchronized to an edge server (NVIDIA Jetson AGX Orin) via a 5G private network every 3 seconds. The raw data is stored in the time series database InfluxDB.

[0151] Model Initialization: The risk decision module loads a dynamic decision tree model with a device health weight of a = 0.3, a service peak weight of b = 0.5, an environmental safety weight of c = 0.2, and a baseline response time of d = 0.8 seconds. The formula for the comprehensive service quality index (S) uses the default parameters α = 0.4, β = 0.3, and γ = 0.3. The environmental baseline value (C_base) is the average of the previous seven days' data (calculated to be 38.6).

[0152] Control group settings: The traditional static threshold strategy is used, the upper limit of the response time is fixed at 4 seconds, the basic caching mechanism is activated when the network load is >85%, and the response function is not interfered with by the environment.

[0153] During the test period, the system will run at full capacity from 9:00 AM to 5:00 PM daily. The risk decision module will execute the following process every 15 minutes:

[0154] 1) The device monitoring unit counts the number of times the ATM vibration exceeds the standard (>0.3g) and the number of times the banknote balance is less than 20% alarm, and generates the device health level D;

[0155] 2) The data acquisition unit integrates the temperature and humidity fluctuation values ​​(hourly change rate) and the peak value of human traffic density to calculate the environmental interference factor C;

[0156] 3) The decision tree inputs D, real-time request volume P (number of queries per minute), and environmental security score E to generate a response time prediction value;

[0157] 4) When the predicted T > dynamic T_max (measured T_max = 2.8 seconds on the 10th day), the policy allocation unit diverts 40% of the queries to edge nodes within 3 kilometers;

[0158] 5) When the network load factor N>0.8, the three-level cache mechanism is activated, and when the environmental interference factor suddenly changes>35%, the encryption protocol is automatically upgraded to AES-256.

[0159] The policy feedback module recorded execution data hourly and optimized the weight coefficients (a→0.25, b→0.55) based on 100,000 pieces of feedback data on the 15th day. During the test, a total of 1.42 million query requests were processed and 260 million pieces of device health data were collected.

[0160] Experimental data record sheet

[0161] Table 1: Comparison of equipment health monitoring (unit: times / day)

[0162]

[0163] Table 2: Business Request Processing Capacity (Average during Peak Hours 11:00-13:00)

[0164]

[0165] Table 3: Environmental disturbance response records (heavy rainfall event on the 22nd day)

[0166]

[0167] Table 4: Service response time distribution (unit: seconds)

[0168]

[0169] Table 5: Comprehensive service quality index (S value ≥ 0.8 is excellent)

[0170]

[0171]

[0172] Table 6: System resource consumption comparison

[0173]

[0174] Data analysis and demonstration of beneficial effects:

[0175] Table 1's device health data reveals the core strength of this system: through real-time monitoring and dynamic decision-making, the experimental group reduced equipment failure alarms by 82.1% (an average of 0.25 vibration exceeding the standard per day compared to 1.95 per day in the control group). Specifically, in terms of CPU temperature management, the deployment of dedicated sensors reduced the probability of overheating in the experimental group to 0.1 per day, significantly lower than the control group's 1.4 per day. This difference stems from the closed-loop optimization mechanism of the policy feedback module described in this system: when Service Terminal 3's temperature was detected to have consistently exceeded the standard on the 12th day, the risk decision module automatically reduced its load weight and migrated query tasks to healthy devices.

[0176] The comparison of business processing capabilities in Table 2 highlights the value of innovative design. Under the premise of a 17.5% increase in the average daily request volume, the response time of the experimental group decreased by 47.6%. This effect is directly attributed to the dynamic strategy coordination mechanism: when the network load coefficient N>0.8 (the measured probability of occurrence is 28.4%), the optimization unit synchronously executes cache priority and edge node diversion. As shown in Table 3 for the rainstorm event, when the environmental interference factor C suddenly changed to 67.8 at 11:15 (an increase of 66.9%), the system simultaneously triggered encryption upgrades and edge node activation to ensure service continuity. In contrast, due to the lack of environmental responsiveness, the control group suffered service interruptions of more than 30 minutes under the same conditions.

[0177] The improved service stability is quantified in Table 4. The P95 response time in the experimental group remained stable at under 2.69 seconds, 2.04 seconds shorter than the control group. The "IoT-optimized query model" effectively reduced long-tail latency. This advantage stems from three innovations: 1) The S-index formula integrates multiple parameters (T / N / C) to provide a unified metric for decision-making; 2) a response time prediction formula enables pre-emptive resource allocation; and 3) the terminal interaction module's voice command parsing capability reduces manual intervention time by 45%.

[0178] Table 5 shows the improvement in system robustness through the comprehensive service quality index. The experimental group's S-index averaged 0.88 (fluctuation range ±0.05), with a volatility of only one-third that of the control group. Especially during the environmental disturbance period (12:00 period in Table 3), when the C-value rose to 83.5, the S-index remained at 0.81; while the control group experienced service degradation under the same conditions (simulated S-value 0.41). More importantly, the system's natural language parsing function reduced voice query response time to 2.1 seconds (basic keyboard and mouse operation: 2.9 seconds), contributing approximately 17% to the improvement in the S-index.

[0179] Table 6 shows resource consumption data validating the benefits of this innovation: the experimental group reduced communication bandwidth by 41.5% and storage IOPS by 36.8%. This benefit stems from two innovative approaches: 1) the temperature and humidity linkage mechanism in the environmental sensing module automatically reduces data collection frequency when the environment is stable; and 2) the dynamic encryption strategy, which enables a lightweight encryption algorithm when C < 40, reducing CPU load by 32%. Compared to the limitations of redundant data collection and single encryption in traditional systems, this solution achieves a balance between resource conservation and service optimization.

[0180] In summary, this embodiment uses six-dimensional data to prove the innovative value of the cluster: in real financial scenarios, the deep coupling of the IoT sensor network and the dynamic decision-making model enables the system to improve service efficiency (response time↓

[0181] Breakthrough improvements were achieved in metrics such as performance (47.6%), stability (S-index ↑49.2%), and resource efficiency (bandwidth ↓41.5%). In particular, in response to the requirements of the "Financial Information System Continuity Specification," the experimental group's service interruption duration was reduced to near zero, providing a new technological paradigm for high-availability architecture in the financial industry.

[0182] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A service capability query method based on the Internet of Things, characterized in that: The method comprises: An IoT sensor array and a data collection terminal are set up outside the financial service facility, and real-time environment and equipment status data are collected based on the IoT sensor array and the data collection terminal to establish a service capability query model; Based on the established service capability query model, real-time data is put into the risk assessment model to determine the optimal service response time parameters; After selecting the optimal service response time parameters, the real-time data and the optimal parameters are put into the query optimization model to determine the optimal query strategy; The service capability query model is optimized based on the preferred service response time parameter and the preferred query strategy, the data collection terminal is adjusted based on the optimized model, and a preset risk monitoring and early warning system is established.

2. The service capability query method based on the Internet of Things according to claim 1, characterized in that: The service capability query model includes: Comprehensive service quality index algorithm: Where α, β, and γ are weight factors, ranging from [0 to 1], and satisfying α + β + γ = 1. The initial value of the weight is preset by the system. S is the comprehensive service quality index, and T is the service response time. max Refers to the maximum threshold of service response time, N is the network load factor, C is the environmental interference factor, C base It is the baseline value of the environmental interference factor and the steady-state reference value set based on historical data.

3. The service capability query method based on the Internet of Things according to claim 1, characterized in that: The IoT sensor array includes: Vibration sensors and cash box status sensors deployed in ATM equipment; CPU temperature sensor and network delay detection module installed on the service terminal; Temperature and humidity sensors and crowd density cameras installed in business premises.

4. The service capability query method based on the Internet of Things according to claim 1, characterized in that: The risk assessment model is a dynamic decision tree model, and its input layer includes: Equipment failure warning level based on equipment health classification; Real-time business request peaks from service queue monitoring; Environmental safety score composed of environmental interference and security system; When determining the optimal service response time parameter, the service response time prediction formula is used: T=a×D+b×N+c×C+d Where D represents the device health indicator, a, b, and c are weight coefficients obtained through historical training, and d is the baseline response time constant.

5. The service capability query method based on the Internet of Things according to claim 1, characterized in that: Determination of the preferred query strategy specifically includes: When T>threshold T_max, distributed edge nodes are enabled to offload query requests; When the network load factor N>0.8, the query result cache priority mechanism is triggered; Based on the mutation amplitude of the environmental interference factor C, the data encryption level is dynamically adjusted.

6. The network management center system based on the Internet of Things is characterized by: It includes an environment perception module (100), a risk decision module (200), a query execution module (300), a terminal interaction module (400) and a strategy feedback module (500); The environment perception module (100) includes a device monitoring unit (101) and a data acquisition unit (102). The device monitoring unit (101) is used to obtain hardware status data of ATM devices and service terminals in real time. The data acquisition unit (102) collects environment and network load data through temperature and humidity sensors and network probes. The risk decision module (200) dynamically optimizes the service response time parameter through the risk assessment model, and transmits the parameter instruction to the query execution module (300); The terminal interaction module (400) allows a user to submit a service capability query request via a mobile terminal or a voice interface, and transmits the request to the query execution module (300); The query execution module (300) includes a strategy allocation unit (301) and an optimization unit (302), wherein the strategy allocation unit (301) selects a query path based on an output instruction of the risk decision module (200), and the optimization unit (302) dynamically adjusts resource allocation according to the prediction formula; The policy feedback module (500) records the execution status of all query requests, summarizes the response delay data and environmental variables, and feeds them back to the risk decision module (200) for model iterative optimization.

7. The network management center system based on the Internet of Things according to claim 6, characterized in that: The equipment monitoring unit (101) integrates a vibration sensor, a banknote box status sensor and a CPU temperature sensor; The data acquisition unit (102) is equipped with an electromagnetic interference detector and a three-dimensional anemometer; The terminal interaction module (300) supports natural language analysis and converts voice requests into structured query instructions.

8. The network management center system based on the Internet of Things according to claim 6, characterized in that: The operation of the optimization unit (302) includes: When the predicted response time T> threshold T max When , the distributed edge nodes are activated to divert query requests; Dynamically switch the data encryption protocol level based on the numerical range of the environmental interference factor; A compensation strategy for the network load factor is generated through the dynamic decision tree model.

9. An electronic device comprising: one or more processors; a storage device having one or more programs stored thereon; When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 5.

10. A computer-readable storage medium having executable instructions stored thereon, which, when executed by a processor, causes the processor to implement the method according to any one of claims 1 to 5.

Images