Intelligent soil sensor for industrial planting

Through the processor chip and built-in algorithm model, the hardware structure of traditional soil sensors is simplified and soil parameters are directly collected, solving the problems of high cost and complex design of traditional soil sensors, and realizing intelligent soil sensors with structure simplification, cost reduction and error control.

CN222070616UActive Publication Date: 2024-11-26ZHENGZHOU UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202420215621.0
Authority / Receiving Office
CN · China
Patent Type
Utility models(China)
Current Assignee / Owner
Filing Date
2024-01-30
Publication Date
2024-11-26
Estimated Expiration
2034-01-30

AI Technical Summary

Technical Problem

The conversion links of traditional soil sensors are too long and have many steps, resulting in high costs, complex hardware design, and difficult to meet the needs of industrial planting.

Method used

Using processor chips combined with built-in algorithm models, a simplified hardware structure is designed, and soil parameters are directly collected through multi-layer perceptron models, reducing circuit conversion steps and power consumption, increasing the number of contact points, using cheap chips and passing through the model in Flash Configuration enables flexible functions.

Benefits of technology

It has achieved structural simplification, cost reduction, and error control within a controllable range, reducing design difficulty and manufacturing costs, and the cost reduction is about one-third of traditional sensors with the same accuracy, and hardware design optimization reduces chip selection requirements.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN222070616U_ABST
    Figure CN222070616U_ABST
Patent Text Reader

Abstract

The utility model provides an intelligent soil sensor for industrial planting. The intelligent soil sensor comprises a sensor shell, an integrated circuit and a plurality of over-current combs, a processor chip, a peripheral circuit and a plurality of contact points are integrated in the integrated circuit, and the plurality of contact points are distributed at the lower end of the integrated circuit; the integrated circuit is installed in the sensor shell, a row of over-current combs are arranged at the bottom end of the shell, and contact points at the lower end of the integrated circuit are electrically connected with the over-current combs in a one-to-one correspondence mode. The intelligent soil sensor has the advantages of low cost, simple structure, flexible configuration and the like, and is suitable for detecting soil indexes in a large-scale industrial crop planting scene.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The utility model relates to the technical field of sensors, and in particular to an intelligent soil sensor for industrial planting. Background Art

[0002] The acquisition idea of ​​traditional soil sensors can be roughly summarized as follows: obtain a physical quantity A, then convert it through physical or chemical means to obtain physical quantity B, and then continue to convert physical quantity B to obtain C. After multiple conversions, the physical quantity N that can be expressed by voltage or resistance is obtained. After sampling and conversion by an AD converter, the final readable value is formed, and finally the physical quantity A is expressed through the corresponding calculation formula.

[0003] There are at least two problems. The first is that the conversion link is too long, resulting in out-of-control costs. The second is that there are too many conversion steps, which leads to complex hardware design. The specific manifestations are mainly reflected in circuit design and production. The high cost and high difficulty have formed corresponding technical barriers.

[0004] With the development of artificial intelligence and the maturity of edge computing theory, there have been various studies in the field of artificial intelligence to solve this problem.

[0005] The core idea is to confirm through a large number of experiments that there is a correlation between the sampled signal during overpowering and the final required parameters. Based on the collected data, a fitting and available multi-layer perceptron model is trained. After quantization and compression of the model, it is deployed on the edge sensor. This design avoids the complicated physical conversion process and can directly obtain the required physical quantity.

[0006] The above research includes "Design and Experiment of High- and Low-Frequency Dual-Frequency Excitation Soil Moisture Content Sensor", "Study on the Impact and Correction of High-Salt Soil Environment on Soil Moisture Sensor", "Study on Calibration of Capacitive Soil Moisture Sensor Based on Random Forest", etc., which provide a theoretical basis for the development of effective and available intelligent solutions for soil sensors.

[0007] However, these studies are all based on principles and methods. The physical architecture they are based on is very different from traditional soil sensors and has its own unique requirements.

[0008] In order to solve the above problems, people have been seeking an ideal technical solution. Utility Model Content

[0009] The purpose of the utility model is to address the deficiencies of the prior art and thus provide an intelligent soil sensor for industrial planting with a simplified structure, better power consumption control, error within a controllable range, and greatly reduced cost.

[0010] In order to achieve the above-mentioned purpose, the technical solution adopted by the utility model is: an intelligent soil sensor for industrial planting, including a sensor housing, an integrated circuit and a plurality of electric combs;

[0011] The integrated circuit integrates a processor chip, a peripheral circuit and a plurality of contact points, and the plurality of contact points are distributed at the lower end of the integrated circuit;

[0012] The integrated circuit is installed in the sensor housing. A row of electric combs is arranged at the bottom end of the housing. The contact points at the lower end of the integrated circuit are electrically connected to the electric combs one by one.

[0013] Based on the above, the processor chip is a single chip microcomputer.

[0014] Based on the above, the single chip microcomputer is connected to a plurality of capacitors through a plurality of IO interfaces, the specification of the capacitors is between 10nf and 200nf, and the other end of each capacitor is connected to a contact point to form a collection end contact.

[0015] Based on the above, the single chip microcomputer is directly connected to a plurality of contact points through a plurality of IO interfaces, which serve as high level output terminals.

[0016] Based on the above, the electric combs are arranged at equal intervals.

[0017] Based on the above, the contact points are arranged at equal intervals on the front and back sides of the integrated circuit.

[0018] Based on the above, the shell is in a wedge-shaped structure with a narrowed lower end.

[0019] The utility model has substantial characteristics and progress compared with the prior art. Specifically, the utility model has the following advantages:

[0020] 1. Since this solution abandons the traditional idea of ​​converting physical quantities multiple times through hardware circuits, it adopts the idea of ​​combining processor chips with built-in algorithm models to design a new adaptive hardware structure, which is converted into a model of chips, peripheral circuits and contact points, which greatly simplifies the circuit structure, reduces the design difficulty, and reduces production and manufacturing costs.

[0021] 2. In order to reduce costs, the number of contact points of soil sensors has been greatly increased, which further reduces the difficulty of sampling and power consumption. It is no longer necessary to use an ADC circuit to measure the required parameters. From the perspective of manufacturing cost, use cost and service life, the overall cost reduction is large, about one-third of the traditional sensors with similar accuracy. At the same time, the optimization of hardware design reduces the requirements for chip selection. As long as the chips with the same package can be replaced by each other, cheap chips can be used in production, which further reduces costs.

[0022] 3. Flexible configuration. You can implement different functions by modifying the weight parameters of the multi-layer perceptron in Flash. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 The utility model discloses a circuit principle diagram of an intelligent soil sensor for industrial planting in the utility model.

[0024] Figure 2 It is a relationship diagram of an integrated circuit board of an intelligent soil sensor for industrial planting in the utility model.

[0025] Figure 3 This is one of the structural schematic diagrams of an intelligent soil sensor for industrial planting in the utility model.

[0026] Figure 4 This is the second structural schematic diagram of an intelligent soil sensor for industrial planting in the utility model.

[0027] In the figure, 20. sensor housing; 21. comb B side; 22. comb A side;

[0028] 1.NET1; 2.NET2; 3.NET3; 4. NET4; 5. NET5; 6. NET6; 7. NET7; 8. NET8; 9. NET9; 10. NET10; 11.NET11; 12. NET12; 13. NET13; 14. NET14; 15. NET15; 16. NET16. DETAILED DESCRIPTION

[0029] The technical solution of the utility model is further described in detail below through specific implementation methods.

[0030] In this embodiment, the processor chip is a single-chip microcomputer, and the AT32 model is taken as an example to illustrate the details of the hardware design.

[0031] like Figure 1-Figure 4 As shown, an intelligent soil sensor for industrial planting comprises a sensor housing (20), an integrated circuit and a plurality of electric combs, wherein the housing (20) is in a wedge-shaped structure with a narrowed lower end.

[0032] The integrated circuit integrates a processor chip, a peripheral circuit and a plurality of contact points, and the plurality of contact points are evenly spaced and distributed front to back at the lower end of the integrated circuit;

[0033] The integrated circuit is installed in the sensor housing (20); a row of equally spaced electric combs are provided at the bottom end of the housing (20); the contact points at the bottom end of the integrated circuit are electrically connected to the electric combs one by one; specifically, the electric combs include an electric comb A side (22) and an electric comb B side (21).

[0034] The single chip microcomputer is connected through several IO interfaces (with Figure 1 Taking the PA0, PA1, PA2, PA3, PA4, PA5, PA6, and PA7 interfaces in the example, several capacitors are connected respectively, and the specification of the capacitor is 100nf. The AT32 microcontroller connects 100nf capacitors C23, C24, C20, C21, C22, C17, C18, and C19 through PA0, PA1, PA2, PA3, PA4, PA5, PA6, and PA7. The other end of each capacitor is connected to a contact point to form a collection end contact. The corresponding contact points are represented as NET1 (1), NET2 (2), NET3 (3), NET4 (4), NET5 (5), NET6 (6), NET7 (7), and NET8 (8).

[0035] The single chip microcomputer is connected through several IO interfaces (with Figure 1 For example, the PB0, PB1, PB2, PB10, PB11, PB9, PB8, and PB7 interfaces in the example are directly connected to several contact points as high-level output terminals. The AT32 microcontroller connects the contact points NET9 (9), NET10 (10), NET11 (11), NET12 (12), NET13 (13), NET14 (14), NET15 (15), and NET16 (16) through PB0, PB1, PB2, PB10, PB11, PB9, PB8, and PB7. In other embodiments, other types of processor chips can also be used, but considering the cost factor, the chip type with lower procurement and use costs should be selected. It takes about 5 to 10 minutes for AT32 to perform a collection and calculation. The soil parameters will not mutate in a stable industrial planting environment, and the longer calculation interval is still within the acceptable range.

[0036] By setting up wireless modules or wired forms, the INT8 quantified multi-layer perceptron model of soil moisture and soil microbial activity is downloaded to the Flash of sensors at different locations. The sensor is inserted into the soil of the insulated planting box so that the electric comb is in full contact with the soil. After installation, the control system controls the sensor to input the sampling instruction and start sampling.

[0037] PB0, PB1, PB2, PB10, PB11, PB9, PB8, and PB7 output high level, and start recording the pulse times of the PWM square wave input by PB6.

[0038] After the soil charge is saturated, capacitors C23, C24, C20, C21, C22, C17, C18, and C19 begin to charge.

[0039] As the capacitor charges, PA0, PA1, PA2, PA3, PA4, PA5, PA6, and PA7 become high level. When reaching the high level, the number of PWM pulses is recorded as , , , , , , , , forming the vector .

[0040] Read the multi-layer perceptron model that has been quantized by INT8 in Flash. Input the model and get the result.

[0041] The output result is mapped to the specified range, and the mapped value is used as the final sampling result. By adjusting the weights of the multi-layer perception model, different types of outputs can be obtained, such as soil moisture, soil ion concentration, soil microbial activity, etc.

[0042] The multilayer perceptron model quantized by INT8 is any verified and available model currently available. The process of obtaining different types of outputs by adjusting the weights of the multilayer perceptron model is also a conventional usage of the multilayer perceptron model. Since it is not the content to be protected by this patent, it will not be elaborated. This application aims to provide an adaptive hardware carrier for the application of this model.

[0043] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the utility model rather than to limit it. Although the utility model has been described in detail with reference to the preferred embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the utility model can still be modified or some technical features can be replaced by equivalents without departing from the spirit of the technical solution of the utility model, which should be included in the scope of the technical solution for protection of the utility model.

Claims

1. An intelligent soil sensor for industrial planting, characterized in that: It includes a sensor housing, an integrated circuit and a plurality of electric combs; The integrated circuit integrates a processor chip, a peripheral circuit and a plurality of contact points, and the plurality of contact points are distributed at the lower end of the integrated circuit; The integrated circuit is installed in the sensor housing. A row of electric combs is arranged at the bottom end of the housing. The contact points at the lower end of the integrated circuit are electrically connected to the electric combs one by one.

2. The intelligent soil sensor for industrial planting according to claim 1, characterized in that: The processor chip is connected to a plurality of capacitors through a plurality of IO interfaces, wherein the specification of the capacitors is between 10nf and 200nf, and the other end of each capacitor is connected to a contact point to form a collection end contact.

3. The intelligent soil sensor for industrial planting according to claim 2, characterized in that: The processor chip is directly connected to a plurality of contact points through a plurality of IO interfaces, respectively, as high-level output terminals.

4. The intelligent soil sensor for industrial planting according to claim 2 or 3, characterized in that: The electric combs are arranged at equal intervals, and the intervals are less than 5 mm and greater than 0.8 mm.

5. The intelligent soil sensor for industrial planting according to claim 4, characterized in that: The contact points are arranged at equal intervals on the front and back sides of the integrated circuit.

6. The intelligent soil sensor for industrial planting according to claim 5, characterized in that: The shell is in a wedge-shaped structure with a narrowed lower end.