Java使用Milo实现OPC UA客户端,封装spring boot starter

一、milo库

由eclipse开源,地址:https://github.com/eclipse/milo,可以基于此开发OPC UA客户端或者服务端。

本文介绍基于milo 封装的spring boot starter,yml配置OPC UA地址,是否匿名等信息,即可连接OPC UA服务器。

二、OPC UA服务端

这里使用的是kepware 6.6 版本,可参考之前的文章配置:Kepware配置OPC UA实现匿名or用户名/密码连接

三、工具使用

3.1 依赖

引入maven仓库地址,当前最新为:3.0.3 ,代码已上传至GitHub,有问题issues,欢迎star

查看最新版本:https://central.sonatype.com/artifact/com.kangaroohy/milo-spring-boot-starter

1
2
3
4
5
<dependency>
<groupId>com.kangaroohy</groupId>
<artifactId>milo-spring-boot-starter</artifactId>
<version>${lastVersion}</version>
</dependency>

3.2 配置

yml配置OPC UA地址,是否匿名等信息

1
2
3
4
kangaroohy:
milo:
endpoint: opc.tcp://127.0.0.1:49320
security-policy: none
1
2
3
4
5
6
kangaroohy:
milo:
endpoint: opc.tcp://127.0.0.1:49320
security-policy: basic256sha256
username: OPCUA
password: 123456

特别提醒:

在kepware中,用户名/密码访问时,opcua配置,安全策略中三个策略全部勾选

同时kepware选项属性中的OPC UA配置,不允许匿名访问

此时,security-policy可选值:basic256sha256,basic256,basic128rsa15都可

同时配置上 用户名/密码 即可访问服务器

3.3 连接池

由于kepware中OPC UA最多只能有128个连接,且milo创建、释放连接比较耗时,因此本工具封装自带了一个连接池配置,默认会生成3个连接,可配置如下信息,达到连接管理的目的

1
2
3
4
5
6
7
kangaroohy:
milo:
pool:
max-idle: 5
max-total: 20
min-idle: 2
initial-size: 3

3.4 写

注入MiloService即可使用,支持:批量读、单个写、批量写、遍历节点信息等

其中:写值时可能需要指定数据类型,视点位情况而定

Opc后边的字段对应Kepware中的tag数据类型(Ua除外,为通用类型)

在这里插入图片描述

id 支持字符串表达式:ns=<命名空间索引>;<标识符类型>=<标识符>

3.4.1 通用类型

如Kep类型为:Boolean、LLong、Long、String、Float、Double,调用方法:miloService.writeToOpcUa(ReadWriteEntity entity)

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
@SpringBootTest
@RunWith(SpringRunner.class)
public class MiloTest {
@Autowired
MiloService miloService;

@Test
public void writeToOpcUa() {
miloService.writeToOpcUa(
ReadWriteEntity.builder()
.identifier("GA.T1.Boolean")
//Kep中是Boolean类型
.value(true)
.build());
miloService.writeToOpcUa(
ReadWriteEntity.builder()
.identifier("GA.T1.LLong")
//Kep中是LLong类型,即:Int64,Java中的Long类型
.value(1235468L)
.build());
miloService.writeToOpcUa(
ReadWriteEntity.builder()
.identifier("GA.T1.Long")
//Kep中是Long类型,即:Int32,Java中的int类型
.value(123456)
.build());
miloService.writeToOpcUa(
ReadWriteEntity.builder()
.identifier("GA.T1.String")
.value("字符串")
.build());
miloService.writeToOpcUa(
ReadWriteEntity.builder()
.identifier("GA.T1.Float")
//Kep中是Float类型
.value(123.123F)
.build());
miloService.writeToOpcUa(
ReadWriteEntity.builder()
.identifier("GA.T1.Double")
//Kep中是Double类型
.value(123.123)
.build());
}
}

3.4.2 已提供方法的类型

如Kep类型为:Short、Word、Byte、Char,调用方法:miloService.writeToOpcXXX(ReadWriteEntity entity),XXX对应kep类型

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
@SpringBootTest
@RunWith(SpringRunner.class)
public class MiloTest {
@Autowired
MiloService miloService;

@Test
public void writeToOpcUa() {
miloService.writeToOpcShort(
ReadWriteEntity.builder()
.identifier("GA.T1.Short")
//Kep中是Short类型,即:Int16,带符号整数
.value(-123)
.build());
miloService.writeToOpcWord(
ReadWriteEntity.builder()
.identifier("GA.T1.Word")
//Kep中是Word类型,即:UInt16,无符号整数
.value(123)
.build());
miloService.writeToOpcByte(
ReadWriteEntity.builder()
.identifier("GA.BIT_8.Byte")
//Kep中是Byte类型,8位无符号整数
.value(123)
.build());
miloService.writeToOpcChar(
ReadWriteEntity.builder()
.identifier("GA.BIT_8.Char")
//Kep中是Char类型,8位带符号整数
.value(-123)
.build());
}
}

3.4.3 其他类型

其他的数据类型,则需要调用方法:miloService.writeSpecifyType(WriteEntity entity),自行指定转换类型.variant(new Variant(xxx))

new Variant(xxx):

new Variant(String[])

new Variant(Unsigned.ushort(“123”))

….

参数类型具体以标签数据类型为准,例如:

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
@SpringBootTest
@RunWith(SpringRunner.class)
public class MiloTest {
@Autowired
MiloService miloService;

@Test
public void writeToOpcUa() {
UByte[] bytes = new UByte[10];
bytes[0] = UByte.valueOf(1);
bytes[1] = UByte.valueOf(2);
bytes[2] = UByte.valueOf(3);
bytes[3] = UByte.valueOf(4);

miloService.writeSpecifyType(
WriteEntity.builder()
.identifier("GA.BIT_8.Bytes")
//Kep中是Byte Array类型
.variant(new Variant(bytes))
.build());
}
}

3.5 读

在这里插入图片描述

读比较简单,传相应的TAG id数组即可,调用方法:readFromOpcUa(List ids)

id格式:通道名.设备名.TAG

如:GA.T1.T1001R_1

3.6 遍历节点

在这里插入图片描述

可遍历指定节点相关信息

3.7 订阅

这里使用的是实现ApplicationRunner接口,实现在项目启动时,自动订阅相关点位

当点位数值发生改变,则会触发回调,根据回调即可实现相应的逻辑

每新增一个订阅都会长期占用一个opc ua连接,不会释放,订阅支持设置扫描时间,具体可查看方法参数

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
@Component
@Slf4j
public class CustomRunner implements ApplicationRunner {

@Autowired
private MiloService miloService;

@Override
public void run(ApplicationArguments args) throws Exception {
sub();
}

private void sub() throws Exception {
List<String> ids = new ArrayList<>();
ids.add("GA.T1.T1001R");
ids.add("GA.T1.String");
miloService.subscriptionFromOpcUa(ids, (id, value) -> log.info("subscription 点位:{} 订阅到消息:{}", id, value));
}
}